JPWO2019012887A1 - Shape correction apparatus and shape correction method for figure pattern - Google Patents

Abstract

A simulation is performed to form an actual pattern on an actual substrate by lithography based on a unit figure (10). First, the unit graphic (10) is divided into polygons (Fig. (A)), and vectors (Va1 to Vd4) are defined on the individual sides of the polygon (Fig. (B)). The vectors arranged overlapping in the same section are removed, leaving the vectors (V1 to V10) along the outline of the unit figure (10) (Fig. (C)). Normal end points (A to J) are defined at the start position of each vector (V1 to V10), and contour line segments (line segment AB and the like) having both normal ends adjacent to each other are defined (Fig. (D)). An evaluation point is defined at the middle point of each contour line segment, and the amount of deviation of each evaluation point is calculated by simulation. A replenishment end point is additionally defined at the position of the intermediate normal end point (B, C, D, G, H, I) which is not a vertex, and a separate end point is secured for each of the left and right contour line segments. Then, correction is performed by moving each contour line segment in a direction orthogonal to the contour line.

Description

  The present invention relates to a shape correction apparatus and shape correction method for a graphic pattern, and more particularly to a technology for correcting the shape of a graphic pattern formed on a substrate through a lithography process.

  In fields where fine patterning of a specific material layer needs to be performed, such as semiconductor device manufacturing processes, fine patterns are formed on a physical substrate through a lithography process involving drawing using light or an electron beam. A method has been taken to form Usually, a fine mask pattern is designed using a computer, and the resist layer formed on the substrate is exposed based on the data of the mask pattern obtained, developed, and then the remaining resist layer is masked. Etching is performed to form a fine pattern on the substrate.

  However, there is a slight discrepancy between the actual figure pattern actually obtained on the substrate through such a lithography process and the original figure pattern designed on the computer. This is because the lithography process includes the steps of exposure, development, and etching, so the actual figure pattern finally formed on the substrate does not exactly match the original figure pattern used in the exposure step. It is for. In particular, in the exposure step, drawing on the resist layer is performed by light or electron beam. At this time, the exposure region actually drawn on the resist layer by the proximity effect (PE: Proximity Effect) is the original figure It is known that the area is slightly larger than the pattern. In addition, in the etching process, since a loading phenomenon of etching occurs, the shapes of the pattern after development and the pattern after etching are different. It is known that the magnitude of the effect of the loading phenomenon of this etching changes according to the area exposed from the resist layer on the surface of the actual substrate. The proximity effect in the drawing process and the loading phenomenon in the etching process are both phenomena that cause a difference between the shape of the original figure pattern and the shape of the actual figure pattern, but the influence of such a phenomenon The range (scale size) is different for each phenomenon.

  Under these circumstances, in a process of manufacturing a semiconductor device including a lithography process, after designing a desired original figure pattern on a computer, the lithography process using this original figure pattern is simulated on a computer, and an actual substrate is obtained. A procedure is performed to estimate the shape of the real figure pattern that will be formed on top. Then, based on the shape (dimension) of the actual figure pattern obtained as a result of simulation, correction is performed on the shape (dimension) of the original figure pattern as necessary, and the correction figure pattern obtained by this correction is actually used. The lithography process will be performed to produce the actual semiconductor device.

  Therefore, in order to manufacture the final product having the precise pattern as designed, the shape of the real figure pattern is accurately estimated by simulation before actually performing the lithography process, and the appropriate correction for the original figure pattern is performed. Need to Therefore, for example, in Patent Document 1 below, a feature factor characterizing the layout of the original figure pattern and a control factor affecting the dimension of the resist pattern formed on the substrate by the lithography process are used as the input layer. A method of performing high-precision simulation using a neural network is disclosed. Further, Patent Document 2 discloses a method of enhancing the accuracy of simulation using two sets of neural networks, and Patent Document 3 is suitable for extracting feature quantities from a pattern of a photomask. It is disclosed a method of improving simulation accuracy by setting various extraction parameters.

JP, 2008-122929, A JP, 2010-044101, A JP, 2010-156866, A

  As described above, when a real figure pattern is formed on the real substrate by a lithography process based on the original figure pattern, simulation will be performed in advance, and if the original figure pattern is used as it is, it will be formed on the real substrate A process is performed to estimate the shape of a real figure pattern and perform necessary correction on the original figure pattern based on the estimation result to obtain a corrected figure pattern. In order to perform such estimation and correction, various shape correction devices have been proposed, including the devices described in the above-mentioned patent documents.

  In a general shape correction apparatus, an evaluation point is set at a predetermined position on an original figure pattern, and based on a feature amount indicating a feature around the evaluation point, the deviation amount on the actual figure pattern for the evaluation point A corrected figure pattern is created by obtaining (process bias) by simulation and performing correction according to the amount of deviation. Therefore, it is very important to set an appropriate number of evaluation points at appropriate positions on the original graphic pattern. Of course, if the density of the evaluation points is increased to set a large number of evaluation points, the accuracy of the simulation can be improved and more accurate correction can be performed. However, as the number of evaluation points increases, the computational load on the simulation increases, and the time to complete the simulation increases. For this reason, it is desired to perform efficient evaluation point setting that can reduce the calculation load as much as possible while maintaining simulation accuracy sufficient for practical use.

  Therefore, according to the present invention, when performing a simulation necessary for correcting the original figure pattern, a figure pattern capable of performing efficient evaluation point setting capable of reducing the operation load as much as possible while maintaining simulation accuracy sufficient for practical use. It is an object of the present invention to provide a shape correction device of

(1) According to a first aspect of the present invention, in forming a real figure pattern on a real substrate by a lithography process based on the original figure pattern, the original figure pattern is formed so that the real figure pattern matches the original figure pattern. In a figure pattern shape correction apparatus for correcting a shape and creating a corrected figure pattern actually used in a lithography process,
An evaluation point setting unit for setting an evaluation point at a predetermined position on an outline indicating the boundary between the inside and the outside for each unit graphic forming the original graphic pattern;
A feature amount extraction unit for extracting feature amounts indicating features around evaluation points for a source figure pattern;
A bias estimation unit for estimating a process bias indicating an amount of deviation between the position on the original figure pattern of the evaluation point and the position on the actual figure pattern based on the feature amount;
A pattern correction unit that generates a corrected figure pattern by performing correction on the original figure pattern based on the estimated value of the process bias output from the bias estimation unit;
Provide
The evaluation point setting unit
A unit graphic input unit for inputting individual unit graphics consisting of polygons;
A unit figure dividing unit which divides a unit figure by a straight dividing line to form a divided figure composed of a plurality of polygons;
A divided figure vector definition unit which defines divided figure vectors on respective sides of the polygon so that each divided figure makes a round along a predetermined forward direction along the outline of the polygon forming the divided figure. When,
An overlapping segment portion of the searched overlapping vector pairs is searched from the set of divided figure vectors, arranged on overlapping sides of a pair of adjacent polygons adjacent to each other and directed in the opposite direction to each other. And a unit figure vector definition unit that defines a plurality of unit figure vectors arranged along the contour of the unit figure by the remaining vectors.
A normal end point definition unit which defines an end point at each of starting point positions of individual unit graphic vectors;
Between two ordinary end points arranged adjacent to each other on the contour line of each unit graphic, an evaluation point definition unit which defines an evaluation point, and
A replenishment end point definition unit which searches for an intermediate normal end point, which is a normal end point not constituting a polygon vertex, and additionally defines a replenishment end point at the same position as the searched intermediate normal end point for each unit figure;
To have
The pattern correction unit
For each evaluation point, the normal end point or replenishment end point adjacent to one side of the evaluation point along the contour of the unit figure is the first end point, and the normal end point or replenishment end point adjacent to the other side is the second A movement process in which a contour line segment as an end point of is defined, and a contour line segment for each evaluation point is moved in a direction orthogonal to the contour line segment by a correction amount according to the estimated value of the process bias of the evaluation point Contour line segment movement processing unit for performing
If the intermediate normal end point and the replenishment end point, which were at the same position as each other before the movement process, are separated into two different positions after the movement process, an additional process of adding a new contour line segment connecting the positions after each movement process A contour line segment addition processing unit that
When the corner normal end point, which is a normal end point that usually constitutes the vertex of the polygon before the movement processing, is separated into two different positions after the movement processing, the two corner normal ends after separation merge to form a new A contour line segment expansion and contraction processing unit that performs expansion and contraction processing to expand and contract each contour line segment including the corner normal point after separation so as to be a corner normal point;
, And a corrected figure pattern composed of a unit figure whose outline is a set of contour line segments after movement processing, addition processing, and expansion / contraction processing is created.

(2) According to a second aspect of the present invention, in the shape correction device for a graphic pattern according to the first aspect described above,
The unit graphic division unit divides the unit graphic to form a trapezoidal divided graphic.

(3) According to a third aspect of the present invention, in the shape correction device for a graphic pattern according to the first or second aspect described above,
When the unit graphic input unit inputs an elongated strip unit graphic extending along a predetermined longitudinal axis and having linear upper and lower sides parallel to each other,
A plurality of divided figures arranged one-dimensionally along the longitudinal axis by dividing the strip-shaped unit figure by a plurality of dividing lines arranged at predetermined intervals along the longitudinal axis. To form the

(4) According to a fourth aspect of the present invention, in the shape correction device for a graphic pattern according to the first or second aspect described above,
When the unit graphic input unit inputs a unit graphic defined on the XY plane,
A plurality of unit graphic division units arrange a plurality of horizontal division lines parallel to the X axis at predetermined intervals in the Y axis direction, and divide the unit graphic by the plurality of horizontal division lines, thereby extending a plurality of elongated shapes in the X axis direction By forming a plurality of strip-shaped figures and further dividing each of the plurality of strip-shaped figures by a plurality of vertical division lines disposed at predetermined intervals along the X axis and parallel to the Y axis, It is intended to form divided figures of.

(5) According to a fifth aspect of the present invention, in the shape correction device of a graphic pattern according to the first to fourth aspects described above,
The unit figure dividing unit has a function of setting the beam maximum forming size of the drawing apparatus used in the lithography process, and sets the arrangement interval of the dividing lines used for dividing the unit figure to the beam maximum forming size or less. is there.

(6) According to a sixth aspect of the present invention, in the shape correction device for a graphic pattern according to the first to fifth aspects described above,
The divided figure vector definition unit numbers the first vertex to the Kth vertex in the order of clockwise or counterclockwise along the outline to each vertex of the divided figure consisting of K square, and the k th vertex By repeatedly performing the process of defining the kth divided figure vector having the (k + 1) th vertex as the start point and the (k + 1) th vertex as the end point (where k = K, the first vertex) up to k = 1 to K The division figure vector of the division figure is defined.

(7) According to a seventh aspect of the present invention, in the shape correction device of a graphic pattern according to the first to sixth aspects described above,
In the evaluation point definition unit, the evaluation point is defined at the middle point position of two normal end points disposed adjacent to each other on the contour line of the unit graphic.

(8) According to an eighth aspect of the present invention, in the shape correction device of a graphic pattern according to the first to seventh aspects described above,
The replenishment end point definition unit defines a total of N normal end points defined for one outline, in the order along the outline, from the first normal end point T (1) to the Nth normal end point T (N ), And focusing on the n-th normal endpoint T (n), the (n-1) -th normal endpoint T (n-1) (where n = 1, the N-th normal endpoint Three points of T (N), the n-th normal end point T (n), and the (n + 1) -th normal end point T (n + 1) (where n = N, the first normal end point T (1)) If it is on a straight line, the n-th normal end point T (n) is determined to be an intermediate normal end point, and the process of additionally defining the replenishment end point at the same position as the intermediate normal end point is repeatedly executed to n = 1 to N. It is something like that.

(9) According to a ninth aspect of the present invention, in the shape correction device for graphic patterns according to the first to eighth aspects described above,
When the contour line segment expansion / contraction unit separates one corner normal end point into a first moving end point and a second moving end point by moving processing, the first moving end point is one end point. The intersection of the straight line including the post-movement contour line segment and the straight line including the second post-movement contour line segment having the second moving end point as one end point is determined, and the intersection point is set as a new corner normal end point, The extension and contraction processing is performed such that one end point becomes the new corner normal end point with respect to the after-moving outline line segment and the second after-moving outline line segment.

(10) According to a tenth aspect of the present invention, in the shape correction device of a graphic pattern according to the first to ninth aspects described above,
The evaluation point setting unit further includes an end point reconstruction unit for performing reconstruction to correct the number or position of the normal end points defined on the outline of the unit graphic by the normal end point definition unit;
The evaluation point definition unit defines an evaluation point based on the normal end points obtained by the above reconstruction.

(11) According to an eleventh aspect of the present invention, in the shape correction apparatus for a graphic pattern according to the tenth aspect described above,
The end point reconstructing unit is configured to perform normal end point reconstruction by an end point removing process of removing some or all of the intermediate normal end points for some or all unit graphics.

(12) A twelfth aspect of the present invention is the shape correction apparatus for a graphic pattern according to the eleventh aspect described above,
The end point reconstructing unit is configured to arrange a total of N normal end points defined for one outline of one unit graphic in the order along the outline, respectively, from the first normal end point T (1) to the Nth Focusing on the n-th normal end point T (n), the n-th normal end point T (n) and a normal end point T (n) that is not to be removed and located immediately before the n-th normal end point T (n) n−m) (where m is a natural number, and in the case of n−m <1, the distance to T (n−m + N)) is L1, and the n-th normal end point T (n) and the (n + 1) th When the distance between the normal end point T (n + 1) of (1) (where n = N, the first normal end point T (1): the same applies hereinafter) is L2
At least one of L1 and L2 is equal to or less than a predetermined reference value Wmerge, and the normal end point T (n-1) (where n = 1, the Nth normal end point T (N)), T (n) , And T (n + 1) are three straight lines, the determination process for the n-th normal end point T (n) to be removed is repeatedly executed until n = 1 to N, and the removal process is normally performed. An end point removal process for removing end points is performed.

(13) According to a thirteenth aspect of the present invention, in the shape correction apparatus for a graphic pattern according to the eleventh aspect described above,
The end point reconstructing unit is configured to arrange a total of N normal end points defined for one outline of one unit graphic in the order along the outline, respectively, from the first normal end point T (1) to the Nth Of the n-th normal end point T (n) and the normal end point T (n-m) not to be removed located immediately before the n-th normal end point T (n), where m is a natural number, In the case of n−m <1, the distance to T (n−m + N) is L1, and the nth ordinary end point T (n) and the (n + 1) th ordinary end point T (n + 1) (where n = N) In this case, let L2 be the distance between the first normal end point T (1) and so on), and the (n + 1) th normal end point T (n + 1) and the (n + 2) normal end point T (n + 2) , N = N−1, the distance between the first normal end point T (1), and n = N, the second normal end point T (2): When 3 and it was,
L2 is less than or equal to a predetermined reference value Wmerge, and the normal end point T (n-1) (where n = 1, the Nth normal end point T (N): the same applies hereinafter), T (n), T When the four points (n + 1) and T (n + 2) lie on a straight line, the normal end point T (n) is targeted for removal if L1 <L3, and the normal end point T (n + 1) if L1> L3. If L1 = L3, then either normal endpoint T (n) or normal endpoint T (n + 1) is targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n-1), T (n) and T (n + 1) are on a straight line, but the normal end point T (n + 2) is If it is not on the line, the end point T (n) is usually targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n), T (n + 1) and T (n + 2) are on a straight line, but the normal end point T (n-1) is a straight line If not on the line, the end point T (n + 1) is usually targeted for removal,
This process is repeatedly performed while increasing n by 1 until n ≧ N is satisfied, and the end point removing process is performed to remove the normal end point to be removed.

(14) A fourteenth aspect of the present invention is the shape correction apparatus for a graphic pattern according to the tenth aspect described above,
The end point reconstruction unit reconstructs the normal end points by an end point rearrangement process in which intermediate normal end points are once removed for some outlines, and new intermediate normal ends are placed on each side of the partial outlines. It is intended to

(15) According to a fifteenth aspect of the present invention, in the graphic pattern shape correction device according to the fourteenth aspect described above,
The end point reconstruction unit performs end point rearrangement processing by repositioning the intermediate normal end points at predetermined intervals on each side of the specific outline after removing the intermediate normal end point for the specific outline. The

(16) According to a sixteenth aspect of the present invention, in the shape correction apparatus for a graphic pattern according to the tenth aspect described above,
The end point reconstruction unit performs end point removal processing by the shape correction device for graphic patterns according to the eleventh to thirteenth aspects described above, and end point rearrangement using the shape correction device for graphic patterns according to the fourteenth or fifteenth aspect described above Processing, and has the function of selectively executing each contour of the unit figure,
End points are reconstructed by the above-mentioned end point removal processing for ordinary endpoints on a regular figure surrounded by unit figure vectors around the forward direction, and are enclosed by unit figure vectors around the reverse direction opposite to the forward direction. With respect to the normal end points on the anti-graphic, end point reconstruction is performed by the above end point rearrangement processing.

(17) According to a seventeenth aspect of the present invention, in the shape correction device for graphic patterns according to the first to sixteenth aspects described above,
The feature amount extraction unit has a function of extracting, for the corrected figure pattern created by the pattern correction unit, a feature amount indicating a feature around an evaluation point on the corrected figure pattern.
The bias estimation unit has a function of estimating a process bias indicating the amount of deviation between the position on the corrected figure pattern of the evaluation point and the position on the actual figure pattern based on the feature amount,
The pattern correction unit has a function of creating a new corrected graphic pattern by further correcting the corrected graphic pattern based on the estimated value of the process bias output from the bias estimation unit,
The correction on the corrected figure pattern is repeatedly performed.

(18) According to an eighteenth aspect of the present invention, in the graphic pattern shape correction device according to the seventeenth aspect described above,
After the contour line segment expansion / contraction unit performs a process of stretching / contracting the contour line segment, a process of correcting the position of the evaluation point with respect to the contour line segment after the stretching process is performed.

  (19) According to a nineteenth aspect of the present invention, an evaluation point setting unit or pattern correction unit included in the shape correction device for graphic patterns or the shape correction device for graphic patterns according to the first to eighteenth aspects described above, It is configured by incorporating a program into a computer.

(20) According to a twentieth aspect of the present invention, in forming a real figure pattern on a real substrate by a lithography process based on the original figure pattern, the real figure pattern matches the original figure pattern, In a shape correction method of a figure pattern for correcting a shape and creating a corrected figure pattern actually used in a lithography process,
An evaluation point setting step in which the computer sets an evaluation point at a predetermined position on an outline indicating the boundary between the inside and the outside for each unit graphic forming the original graphic pattern;
A feature amount extraction stage in which a computer extracts feature amounts indicating features around evaluation points for the original figure pattern;
A bias estimation step in which a computer estimates a process bias indicating a deviation between a position on an original figure pattern of an evaluation point and a position on a real figure pattern based on the feature amount;
A pattern correction step in which a computer generates a corrected figure pattern by correcting the original figure pattern based on the estimated value of the process bias obtained by the bias estimation step;
Have
The evaluation point setting stage is
A unit figure input step for inputting individual unit figures composed of polygons;
A unit figure dividing step of dividing a unit figure by a straight dividing line to form a divided figure consisting of a plurality of polygons;
In each divided figure, a divided figure vector defining step for defining divided figure vectors on respective sides of the polygon so as to go around a predetermined forward direction along the outline of the polygon constituting the divided figure. When,
An overlapping segment portion of the searched overlapping vector pairs is searched from the set of divided figure vectors, arranged on overlapping sides of a pair of adjacent polygons adjacent to each other and directed in the opposite direction to each other. And a unit graphic vector defining step of defining a plurality of unit graphic vectors arranged along the contour of the unit graphic by the remaining vectors.
A normal end point defining step that defines an end point at each of starting point positions of individual unit graphic vectors;
An evaluation point defining step of defining an evaluation point between two ordinary end points disposed adjacent to each other on the contour line of the unit graphic;
A replenishment end point defining step of searching for an intermediate normal end point which is a normal end point not constituting a polygon vertex for each unit figure, and additionally defining a replenishment end point at the same position as the searched intermediate normal end point;
Have
The pattern correction stage is
For each evaluation point, the normal end point or replenishment end point adjacent to one side of the evaluation point along the contour of the unit figure is the first end point, and the normal end point or replenishment end point adjacent to the other side is the second A movement process in which a contour line segment as an end point of is defined, and a contour line segment for each evaluation point is moved in a direction orthogonal to the contour line segment by a correction amount according to the estimated value of the process bias of the evaluation point Contour line segment moving step to
If the intermediate normal end point and the replenishment end point, which were at the same position as each other before the movement process, are separated into two different positions after the movement process, an additional process of adding a new contour line segment connecting the positions after each movement process Contour line segment adding step to
When the corner normal end point, which is a normal end point that usually constitutes the vertex of the polygon before the movement processing, is separated into two different positions after the movement processing, the two corner normal ends after separation merge to form a new An outline line segment expansion / contraction step for expanding and contracting each outline line segment including the corner area normal end point after separation so as to be a corner area normal end point;
, And a corrected figure pattern composed of a unit figure whose outline is a set of contour line segments after movement processing, addition processing, and expansion / contraction processing is created.

(21) A twenty-first aspect of the present invention is the shape correction method for a graphic pattern according to the twentieth aspect described above,
The evaluation point setting step further includes an end point reconstruction step for performing a reconstruction that corrects the number or position with respect to the normal end points defined by the normal end point definition step;
In the evaluation point definition step, evaluation points are defined based on the normal end points obtained by reconstruction.

  According to the shape correction apparatus and shape correction method of a figure pattern according to the present invention, after the unit figure constituting the original figure pattern is divided into polygons, end points are defined based on the vertex position of this polygon, An evaluation point is defined for each of the contour line segments connecting two adjacent end points, and the position can be corrected for each of the contour line segments. Therefore, when performing correction on the original graphic pattern, it is possible to perform efficient evaluation point setting that can reduce the calculation load as much as possible while maintaining simulation accuracy sufficient for practical use. Also, if an embodiment is further performed in which the end point is further reconstructed with respect to the end point defined on the basis of the vertex position of the polygon, the number and position of the end point are corrected, and more efficient evaluation point setting is performed. It becomes possible.

FIG. 1 is a block diagram showing a configuration of a figure pattern shape correction apparatus 100 according to a prior application invention and a basic embodiment of the present invention. It is a top view which shows an example in which the difference in shape arose between the original figure pattern and the real figure pattern. In the example shown in FIG. 2, it is a top view which shows the example of a setting of each evaluation point, and the process bias which arises in each evaluation point. It is a flowchart which shows the design and manufacturing process of a product using the shape correction apparatus 100 of the figure pattern shown in FIG. It is a top view which shows the concept which grasps | ascertains the feature of the periphery about the evaluation point defined on the outline of a figure pattern. It is a figure which shows the outline | summary of the process performed in the feature-value extraction unit 120 and bias estimation unit 130 which are shown in FIG. It is a flowchart which shows the process sequence performed in the feature-value extraction unit 120 shown in FIG. It is a top view which shows the example of the original figure pattern 10 given to the shape correction apparatus 100 of the figure pattern shown in FIG. It is a top view which shows the state in which the process which superimposes the original figure pattern 10 on the mesh which consists of a two-dimensional array of pixels in the original image creation part 121 shown in FIG. 1 was performed. It is a figure which shows the area density map M1 produced based on the original figure pattern 10 shown in FIG. It is a figure which shows the edge length density map M2 produced based on the original figure pattern 10 shown in FIG. It is a top view which shows the original figure pattern 10 containing the information of dose amount. It is a figure which shows the dose density map M3 created based on the dose-added original figure pattern 10 shown in FIG. It is a top view which shows the procedure which produces the k-th hierarchical image Pk by performing the filter processing using Gaussian filter GF33 to the k-th preparatory image Qk. It is a top view which shows the k-th hierarchy image Pk obtained by the filter process shown in FIG. It is a top view which shows the example of the image processing filter utilized for the filter processing shown in FIG. It is a top view which shows another example of the image processing filter utilized for the filter processing shown in FIG. It is a top view which shows the procedure which produces the (k + 1) th preparatory image Q (k + 1) by performing an average pooling process with respect to the k-th hierarchy image Pk. It is a top view which shows the procedure which produces the (k + 1) th preparatory image Q (k + 1) by performing a max pooling process with respect to the k-th hierarchy image Pk. It is a top view which shows the procedure which produces | generates the image pyramid PP which consists of n hierarchy images P1-Pn in the image pyramid production | generation part 122 shown in FIG. It is a top view which shows the procedure which calculates the feature-value about the evaluation point E from each hierarchy image in the feature-value calculation part 123 shown in FIG. It is a figure which shows the specific calculation method used by the feature-value calculation procedure shown in FIG. It is a top view which shows the procedure which produces the image pyramid PD which consists of n difference image D1-Dn in the image pyramid production part 122 shown in FIG. It is a block diagram which shows the Example using a neural network as the presumed calculating part 132 shown in FIG. It is a diagram which shows the concrete arithmetic process performed by the neural network shown in FIG. It is a figure which shows the computing equation which calculates | requires each value of the 1st hidden layer in the diagram shown in FIG. FIG. 27 is a view showing a specific example of the activation function f (ξ) shown in FIG. 26. It is a figure which shows the computing equation which calculates | requires each value of the 2nd hidden layer-the Nth hidden layer in the diagram shown in FIG. It is a figure which shows the computing equation which calculates | requires the value y of the output layer in the diagram shown in FIG. It is a flowchart which shows the procedure of the learning phase for obtaining the learning information L which the neural network shown in FIG. 24 uses. It is a flowchart which shows the detailed procedure of the presumed calculating part learning of step S84 in the flowchart shown in FIG. It is a block diagram which shows the structure of the evaluation point setting unit 110 in the shape correction apparatus 100 (refer FIG. 1) of the figure pattern which concerns on basic embodiment of this invention. It is a top view which shows the example of the specific division process performed by the unit figure division part 112 shown in FIG. It is a top view which shows the example of the specific division | segmentation process performed with respect to the parallelogram type | mold unit figure 10 by the unit figure division part 112 shown in FIG. It is a top view which shows the example of the trapezoid divided figure obtained by the division process by the unit figure division part 112 shown in FIG. It is a top view which shows the state which shape correction was performed with respect to each division | segmentation figure shown in FIG.34 (c). FIG. 37 is a plan view showing an example of an exposure process based on the unit figure 15 after correction shown in FIG. 36. An example in which processing by the divided figure vector definition unit 113, the unit figure vector definition unit 114, and the normal end point definition unit 115 shown in FIG. 32 is performed on the parallelogram unit figure 10 shown in FIG. It is a top view shown. A plane showing an example in which processing by the unit figure dividing unit 112, the divided figure vector definition unit 113, the unit figure vector definition unit 114, and the normal end point definition unit 115 shown in FIG. 32 is performed on the L-shaped unit FIG. FIG. 32 shows an example in which processing by unit graphic division unit 112, divided graphic vector definition unit 113, unit graphic vector definition unit 114, and normal end point definition unit 115 shown in FIG. It is a top view. FIG. 33 is a schematic view (FIG. (A)) and a flow chart (FIG. (B)) showing a procedure in the case of adopting a three-point end point removal process as the end point reconstruction process by the end point reconstruction unit 116 shown in FIG. An example in which the end point is reconstructed by applying the end point elimination process of the three-point type shown in FIG. 41 to the normal end points A to J on the parallelogram type unit graphic 10 shown in FIG. It is a top view shown. An example is shown in which the end point is reconstructed by applying the end point removal process of the three-point type shown in FIG. 41 to the normal end points A to I on the L-shaped unit graphic 10 shown in FIG. It is a top view. FIG. 33 is a schematic view (FIG. (A)) and a flow chart (FIG. (B)) showing a procedure in the case of adopting the four-point end point removal process as the end point reconstruction process by the end point reconstruction unit 116 shown in FIG. It is a schematic diagram which shows the procedure at the time of employ | adopting an end point rearrangement process as an end point reconstruction process by the end point reconfiguration | reconstruction part 116 shown in FIG. It is a flowchart which shows the procedure at the time of employ | adopting an end point rearrangement process as an end point reconstruction process by the end point reconstruction part 116 shown in FIG. A plan view showing an example in which the end point rearrangement process shown in FIG. 46 is applied to the normal end points M to T on the square-shaped unit graphic 10 shown in FIG. It is. FIG. 43 is a plan view showing an example in which evaluation points 1 to 8 are defined on the parallelogram type unit graphic 10 after the end point reconstruction shown in FIG. 42 (b) by the evaluation point definition unit 117 shown in FIG. FIG. 43 is a plan view showing an example in which evaluation points 1 to 8 are defined on the L-shaped unit graphic 10 after the end point reconstruction shown in FIG. 43 (b) by the evaluation point definition unit 117 shown in FIG. FIG. 45 is a plan view showing an example in which evaluation points 1 to 18 are defined on the B-shaped unit graphic 10 after the end point reconstruction shown in FIG. 47 (b) by the evaluation point definition unit 117 shown in FIG. FIG. 49 is a plan view showing an example in which replenishment end points are defined on each unit figure shown in FIG. 48 (b), FIG. 49 (b) and FIG. 50 (b) by a replenishment end point definition unit 118 shown in FIG. It is a schematic diagram (figure (a), (b)) and a flowchart (figure (c)) which show the specific procedure of the replenishment end point definition process performed by the replenishment end point definition part 118 shown in FIG. FIG. 2 is a block diagram showing a pattern correction unit 140 in a figure pattern shape correction apparatus 100 (see FIG. 1) according to a basic embodiment of the present invention and components related thereto. FIG. 52 is an enlarged plan view showing each end point and each evaluation point defined on the L-shaped unit graphic shown in FIG. 51 (b). It is a figure which shows an example of the data format showing the interrelationship of each end point and each evaluation point which are shown in FIG. FIG. 55 is a plan view showing a state in which contour line segment movement processing has been performed on an L-shaped unit graphic 10 shown in FIG. 54 by the contour line segment movement processing unit 141 shown in FIG. 53; FIG. 55 is a plan view showing a state in which contour line segment addition processing has been performed on the L-shaped unit graphic after movement processing shown in FIG. 56 by the contour line segment addition processing unit 142 shown in FIG. 53; FIG. 60 is a plan view showing a state in which the contour line segment expansion and contraction processing has been performed on the L-shaped unit graphic after the addition processing shown in FIG. 57 by the contour line segment expansion and contraction processing unit 143 shown in FIG. 58 is a plan view showing a state in which the contour line segment of the L-shaped unit graphic after the stretching process shown in FIG. 58 has been subjected to processing of correcting the position of the evaluation point by the contour line segment stretching processing unit 143 shown in FIG. is there. It is a top view which shows the procedure of the conventional method of defining the end point by predetermined spacing for every edge | side of a unit figure. In the present invention, it is a top view showing the merit which divides a unit figure into a polygon and defines an end point on the basis of the vertex position of this polygon. It is a top view which shows the merit which defines a replenishment end point. In the present invention, it is a top view showing the merit which moves a replenishment end point in the direction which intersects perpendicularly with an outline. It is a top view which shows the demerit at the time of moving an intermediate | middle normal end point and a replenishment end point to arbitrary directions. FIG. 59 is a plan view showing a disadvantage in the case where end point reconstruction is not performed in the example shown in FIG. 58.

  Hereinafter, the present invention will be described based on illustrated embodiments. The technology according to the present invention is applied to the invention described in International Application No. PCT / JP2018 / 022100 (hereinafter referred to as the prior application) based on the Patent Cooperation Treaty (hereinafter referred to as the prior application invention). It has been developed. Therefore, for convenience of explanation, the invention of the prior application will be described first in the following 1〜 1 to 4 4 and the features of the present invention will be described in 5 5 and thereafter. Accordingly, the contents (the contents shown in FIGS. 1 to 31) described in § 1 to 4 below are substantially the same as the basic embodiment described in the prior application.

<<< 1. 1. Basic configuration of shape correction device according to prior application invention >>>
Here, the configuration of the figure pattern shape correction apparatus 100 according to the basic embodiment of the invention of the prior application will be described with reference to the block diagram of FIG. As illustrated, the shape correction apparatus 100 of the figure pattern includes an evaluation point setting unit 110, a feature extraction unit 120, a bias estimation unit 130, and a pattern correction unit 140. Here, three units of the evaluation point setting unit 110, the feature value extraction unit 120, and the bias estimation unit 130 constitute the figure pattern shape estimation apparatus 100 'according to the prior invention, and the figure according to the prior invention The pattern shape correction apparatus 100 is configured by adding a pattern correction unit 140 as a fourth unit to the figure pattern shape estimation apparatus 100 '.

<1.1 Shape estimation device for graphic pattern>
First, the configuration and function of a figure pattern shape estimation apparatus 100 'will be described. The figure pattern shape estimation apparatus 100 ′ serves to estimate the shape of the real figure pattern 20 formed on the real substrate S by simulating a lithography process using the original figure pattern 10. At the top of FIG. 1, an arrow of a dashed dotted line extending rightward from the original figure pattern 10 is shown, and a real substrate S having a figure pattern 20 is drawn at the tip of the arrow. The dashed dotted arrow indicates the physical lithography process.

  The illustrated original figure pattern 10 is data of a figure pattern created by a design operation using a computer, and an arrow of a dashed dotted line indicates a lithography process such as physical exposure, development, etching, etc. based on this data. It shows that the actual substrate S is produced by carrying out. On the real substrate S, a real figure pattern 20 corresponding to the original figure pattern 10 is formed. However, when such a lithography process is performed, a slight discrepancy occurs between the original graphic pattern 10 and the real graphic pattern 20. This is because, as described above, it is difficult to form an accurate figure as the original figure pattern 10 on the actual substrate S due to the conditions of steps of exposure, development, and etching included in the lithography process.

  FIG. 2 is a plan view showing a specific example in which a difference in shape occurs between the original figure pattern 10 and the actual figure pattern 20. As shown in FIG. In a semiconductor device etc., it is actually necessary to form a very fine and complicated figure pattern on the surface of an actual substrate S such as silicon, but here, for convenience of explanation, a simple as shown in FIG. 2 (a) The case where a basic figure is given as the original figure pattern 10 will be described. The original graphic pattern 10 shown in the figure is a pattern consisting of one rectangle, and for example, original graphic data showing formation of a material layer corresponding to the hatched rectangular inner area on the real substrate S. Become.

  In an actual lithography process, a resist layer is formed on the upper surface of the material layer on the actual substrate S, and exposure with light or an electron beam is performed to write on the resist layer. For example, after exposing the resist layer to the inner region (hatched part) of the original figure pattern 10 shown in FIG. 2A, the resist layer is developed to remove the non-exposed part, The exposed portion remains as a remaining resist layer (hatched portion). Furthermore, if the material layer is etched using this remaining resist layer as a mask, theoretically, the internal region (hatched part) of the original figure pattern 10 can be left even for the material layer, On the real substrate S, the same real figure pattern as the original figure pattern 10 shown in FIG. 2A can be obtained.

  However, actually, the actual figure pattern 20 obtained on the actual substrate S does not exactly match the original figure pattern 10. The reason is that the conditions of the steps of exposure, development and etching included in the lithography process affect the shape of the real figure pattern 20 finally obtained. For example, in the exposure step, drawing on the resist layer is performed by light or electron beam, but at that time, the exposure area actually drawn on the resist layer by the proximity effect (PE: Proximity Effect) is the original figure It is known that the area is slightly larger than the pattern 10. Under the influence of such a proximity effect, the real figure pattern 20 obtained on the real substrate S is an area which is wider than the original figure pattern 10 (shown by a broken line) as shown in FIG. 2 (b). become.

  In addition, the characteristics of the developing solution used in the developing process, the characteristics of the etching solution and plasma used in the etching process, and conditions such as the time of the developing process and the etching process also affect the shape of the actual figure pattern 20. Therefore, when actually manufacturing a semiconductor device or the like, after designing a desired original figure pattern 10 on a computer, the lithography process using this original figure pattern 10 is simulated on a computer, and on the actual substrate S A procedure is performed to estimate the shape of the real graphic pattern 20 that will be formed. The figure pattern shape estimation apparatus 100 'shown in FIG. 1 is an apparatus having the function of performing such estimation, and does not actually perform the lithography process (exposure, development, and etching steps) for producing the actual substrate S. In addition, it has a function to estimate the shape of the real figure pattern 20 that will be formed on the real substrate S by simulation.

  In the figure pattern shape estimation apparatus 100 ′ shown in FIG. 1, the evaluation point setting unit 110 sets an evaluation point E on the original graphic pattern 10. Specifically, figure data including information of an outline indicating the boundary between the inside and the outside of the figure is given to the shape estimation apparatus 100 'as the original figure pattern 10, and the evaluation point setting unit 110 Based on the original graphic pattern 10, processing is performed to set an evaluation point at a predetermined position on the outline.

  FIG. 3 is a plan view showing an example in which several evaluation points are set for the original figure pattern 10 shown in FIG. 2A and a process bias (dimension error) generated at each evaluation point. First, FIG. 3A is a plan view showing an example in which evaluation points E11, E12 and E13 are set on the outline of the original figure pattern 10 (rectangular figure) shown in FIG. 2A. Although FIG. 3 shows a simple example in which three evaluation points E11, E12, and E13 are set for convenience of explanation, actually, more evaluation points are set on each side of the rectangle. Ru. For example, it is possible to automatically set a large number of evaluation points by setting the evaluation points continuously at a predetermined pitch along the outline. The present invention proposes a new method for setting this evaluation point, and the specific content thereof will be described in 5 5 and thereafter.

  Here, let us consider the case where an actual figure pattern 20 as shown in FIG. 2 (b) is obtained based on the original figure pattern 10 shown in FIG. 2 (a). FIG. 3 (b) is a plan view showing the contour (solid line) of the real figure pattern 20 shown in FIG. 2 (b) in contrast to the contour (broken line) of the original figure pattern 10 shown in FIG. 2 (a). A state is shown in which the outline of the real figure pattern 20 shown by a solid line extends outward by the dimension y as compared with the outline of the original figure pattern 10 shown by a broken line. Therefore, the width a of the original figure pattern 10 is expanded to the width b of the actual figure pattern 20. The vertical width is also slightly expanded.

  In FIG. 3B, the evaluation points E21, E22 and E23 on the real figure pattern 20 are evaluation points determined as points corresponding to the evaluation points E11, E12 and E13 on the original figure pattern 10. Here, the evaluation point E21 is defined as a point obtained by shifting the evaluation point E11 by the predetermined dimension y11 toward the outer side in the normal direction of the outline indicated by the broken line. Similarly, the evaluation point E22 is defined as a point obtained by shifting the evaluation point E12 by the predetermined dimension y12 toward the outside of the normal line direction of the outline shown by the broken line, and the evaluation point E23 is an outline showing the evaluation point E13 by the broken line. It is defined as a point shifted by a predetermined dimension y13 toward the outside in the normal direction of.

  As in the example shown in FIG. 3, here, in order to quantitatively show the shape change of the real figure pattern 20 with respect to the original figure pattern 10, the deviation amount y in the normal direction of the contour line generated for each evaluation point E is I will use it. In addition, this deviation amount y is called “process bias y” because it is a bias amount generated due to the lithography process. The process bias y is a value having a positive or negative sign, and in the embodiment shown below, the direction in which the exposed portion (drawing portion) of the figure is thick is defined as a positive value, and the narrowing direction is defined as a negative value. In the illustrated example, the inside of the figure surrounded by the outline becomes the exposure part (drawing part), and therefore, when shifted in the outward direction of the outline, a positive value is deviated in the inward direction of the outline. Is a negative value. In the case of the example shown in FIG. 3, since all the evaluation points E deviate in the outward direction, the process biases y11, y12 and y13 take positive values.

  Although only three evaluation points E11, E12, and E13 are shown in FIG. 3 for the sake of convenience of the description, in actuality, more evaluation points E are defined on the outline of the original figure pattern 10. Ru. Therefore, if the process bias y can be estimated for each evaluation point E, the position of each evaluation point E after the lithography process can be determined by shifting each evaluation point E in the normal direction of the contour by the process bias y. The outline position of the real figure pattern 20 can be estimated.

  However, the value of process bias y of each evaluation point E is different for each individual evaluation point. For example, in the case of the example shown in FIG. 3 (b), the values of the process biases y11, y12 and y13 are individual values. This is because the relative positions of the evaluation points E11, E12, and E13 with respect to the original figure pattern 10 are different from each other, and therefore the influence of the lithography process is also different, and a difference also occurs in the amount of deviation generated in each. Therefore, in order to increase the estimation accuracy in estimating the shape of the actual figure pattern 20 by simulation based on the original figure pattern 10, the influence of the lithography process is appropriately predicted for each individual evaluation point, and an appropriate process bias is provided. It is important to find y.

  Therefore, in the figure pattern shape estimation apparatus 100 ′ shown in FIG. 1, an evaluation point is first set on the original graphic pattern 10 by the evaluation point setting unit 110. Specifically, each evaluation point E may be set at a predetermined position on the outline based on the information of the outline indicating the boundary between the inside and the outside of the figure included in the original figure pattern 10. For example, evaluation points can be set continuously at predetermined intervals along the outline (the evaluation point setting method specific to the present invention will be described in § 5 and later).

  Subsequently, the feature amount extraction unit 120 extracts feature amounts indicating features around each evaluation point E for the original figure pattern 10. The feature amount x for one evaluation point E is a value indicating the feature around the evaluation point E. In order to extract such a feature amount x, the feature amount extraction unit 120 includes an original image creation unit 121, an image pyramid creation unit 122, and a feature amount calculation unit 123, as shown in FIG.

  The original image creation unit 121 creates an original image composed of a set of pixels each having a predetermined pixel value based on the given source graphic pattern 10. For example, when the original figure pattern 10 as shown in FIG. 2A is given, the pixel value 1 is for the pixel inside the rectangle (the pixel in the hatched portion in the figure), and the pixel value 0 for the pixel outside the rectangle. To create an original image consisting of a binary image.

  The image pyramid creation unit 122 performs an image pyramid creation process including a reduction process of reducing the original image to create a reduced image, and creates an image pyramid composed of a plurality n of hierarchical images. The n hierarchical images constituting each hierarchy of the image pyramid are all images obtained by performing predetermined image processing on the original image generated by the original image generation unit 121, and have different sizes. There is. Such a collection of hierarchical images is referred to as an “image pyramid” because, when hierarchical images are formed sequentially by stacking hierarchical images from large to small size, it is as if pyramids are formed. It is to see.

  The feature amount calculation unit 123 calculates a feature amount based on the pixel values of the pixels in the vicinity of the evaluation point E for each of the n hierarchical images forming the hierarchy of the image pyramid. Specifically, the feature amount x1 is calculated based on the pixel values of the pixels in the vicinity of the evaluation point E in the first hierarchical image, and the pixel values of the pixels in the vicinity of the evaluation point E in the second hierarchical image are calculated. The feature amount x2 is calculated based on the above, and similarly, the feature amount xn is calculated based on the pixel values of the pixels in the vicinity of the evaluation point E in the n-th hierarchical image. n feature quantities x1 to xn are extracted. For example, in the case of the example shown in FIG. 3A, n feature amounts x1 (E11) to xn (E11) are extracted for the evaluation point E11, and n feature amounts x1 (E12) for the evaluation point E12. .About.xn (E12) are extracted, and n feature amounts x1 (E13) to xn (E13) are extracted for the evaluation point E13.

  On the other hand, the bias estimation unit 130 is a process that indicates the amount of deviation between the position of the evaluation point E on the original figure pattern 10 and the position on the actual figure pattern 20 based on the feature quantity x extracted by the feature quantity extraction unit 120. A process is performed to estimate the bias y. In order to perform such estimation processing, the bias estimation unit 130 includes a feature amount input unit 131 and an estimation operation unit 132. The feature amount input unit 131 is a component to which the feature amounts x1 to xn calculated for the evaluation point E by the feature amount calculation unit 123 are input, and the estimation operation unit 132 performs learning obtained in the learning step performed in advance. Based on the information L, estimated values corresponding to the feature amounts x1 to xn are obtained, and the obtained estimated values are output as an estimated value y of the process bias for the evaluation point E.

  More specifically, for each evaluation point E located on the outline of the figure forming the original figure pattern 10, the estimation calculation unit 132 estimates the process bias value as the amount of deviation in the normal direction of the outline. It will output y. For example, for the original figure pattern 10 shown in FIG. 3A, as shown in FIG. 3B, the process bias y11 for the evaluation point E11, the process bias y12 for the evaluation point E12, and the process bias y13 for the evaluation point E13. Are respectively output from the estimation operation unit 132 as estimated values. Thus, if the estimated value y of the process bias for each evaluation point E is obtained, the new position of each evaluation point E (the position shifted by the process bias y in the normal direction of the contour line) can be determined. Therefore, as shown in FIG. 3 (b), the shape of the real figure pattern 20 can be estimated.

  The above is the basic configuration and basic operation of the figure pattern shape estimation apparatus 100 'according to the invention of the prior application. The specific operation of the feature quantity extraction unit 120 will be described in detail in Section 2, and the specific operation of the bias estimation unit 130 will be described in detail in Section 3.

<1.2 Shape correction device for graphic pattern>
Subsequently, the configuration and function of the figure pattern shape correction apparatus 100 according to the prior invention will be described. The figure pattern shape correction apparatus 100 corrects the shape of the original figure pattern 10 using the figure pattern shape estimation apparatus 100 'described above, and as shown in FIG. 1, the figure pattern shape estimation apparatus 100' A pattern correction unit 140 is further provided in addition to the evaluation point setting unit 110, the feature value extraction unit 120, and the bias estimation unit 130, which are components of The pattern correction unit 140 is a component that performs correction on the original figure pattern 10 based on the estimated value y of the process bias output from the bias estimation unit 130, and a corrected figure pattern obtained by the correction by the pattern correction unit 140. 15 is the final output of the shape correction apparatus 100 of this figure pattern.

  The correction by the pattern correction unit 140 is performed by moving each evaluation point E on the original figure pattern 10 in a direction to offset the process bias y and correcting the boundary of each figure to the position of each evaluation point E after movement. It can be carried out. For example, consider the case where an actual figure pattern 20 as shown in FIG. 3 (b) is estimated for the original figure pattern 10 shown in FIG. 3 (a). In this case, the width a of the rectangle included in the original figure pattern 10 is increased to the width b on the real figure pattern 20, and if (b−a) / 2 = y, the positions of the left and right sides of the rectangle Are all moved inward by y, a rectangle having the same width a as the original graphic pattern 10 is obtained. Therefore, basically, a correction is performed to reduce the horizontal width a of the original figure pattern 10 by 2 y, and the lithography process is performed based on the figure pattern after the correction. , A rectangle having a width a as originally designed can be obtained.

  In the case of the original figure pattern 10 shown in FIG. 3A, the evaluation point E11 is moved to the left (inside the rectangle) by the process bias y11, and the evaluation point E12 is moved to the left (inside the rectangle) by the process bias y12. And the evaluation point E13 may be moved upward (inside the rectangle) by the process bias y13. Of course, in fact, a greater number of evaluation points are defined, so that correction is made to move all these evaluation points by the size corresponding to the process bias to the inside of the rectangle, and new evaluation points are connected. By defining such an outline, a corrected figure pattern 15 including a figure defined by the outline is obtained. Since such correction processing itself is a known technique, detailed description will be omitted here.

  However, actually, even if a lithography process is performed using the corrected figure pattern 15 thus obtained to form a real figure pattern 25 (not shown) on the real substrate S, the real figure pattern 25 obtained is It does not exactly match the original graphic pattern 10 of the original design (for example, the width of the rectangle formed on the real substrate S does not exactly become a). This is because the figures included in the original figure pattern 10 and the figures included in the corrected figure pattern 15 have different sizes and shapes, so that differences occur in the effects such as proximity effect when the lithography process is performed.

  In other words, even if it is found that the process bias y is generated as a result of performing a simulation for a specific evaluation point E, merely moving the position of the evaluation point E in the reverse direction by the process bias y It is not possible to make an accurate correction.

  Of course, compared to the actual figure pattern 20 obtained as a result of performing the lithography process using the original figure pattern 10, the actual figure pattern 25 obtained as a result of performing the lithography process using the corrected figure pattern 15 is Since the pattern becomes closer to the original figure pattern 10, the actual lithography process is performed using the corrected figure pattern 15 obtained by the correction by the pattern correction unit 140 rather than performing the actual lithography process using the original figure pattern 10 as it is. More accurate figure patterns can be obtained on the actual substrate S by executing. That is, if correction is performed by the pattern correction unit 140, it is true that the error is reduced.

  Therefore, practically, as shown in FIG. 1, the process of giving the corrected figure pattern 15 output from the pattern correction unit 140 to the figure pattern shape correction apparatus 100 is repeated again. That is, the corrected figure pattern 15 is given to the figure pattern shape estimation apparatus 100 'as a new original figure pattern, and this new original figure pattern (corrected figure pattern 15) is described in § 1.1. Processing is performed. Specifically, the evaluation point setting unit 110 sets each evaluation point E on the corrected graphic pattern 15, the feature amount extraction unit 120 extracts a feature amount for each evaluation point E, and the bias estimation unit 130 An estimated process bias value y for each evaluation point E is calculated. Then, using the calculated process bias estimated value y, the pattern correction unit 140 performs another correction process.

  As described above, the figure pattern shape correction apparatus 100 shown in FIG. 1 has a function of repeatedly executing correction on the figure pattern. That is, a first corrected figure pattern 15 is obtained based on the original figure pattern 10, a second corrected figure pattern is obtained based on the first corrected figure pattern 15, and a second corrected figure pattern is obtained based on the second corrected figure pattern. Thus, the third corrected figure pattern is obtained, and the correction process of ... is repeated. Each time the correction processing is performed, the shape error between the original graphic pattern and the graphic pattern obtained by simulation is reduced.

  Therefore, for example, correction is completed when the shape error between the original figure pattern and the figure pattern obtained by simulation converges within a predetermined tolerance, and the actual lithography process is executed using the finally obtained corrected figure pattern. Then, on the real substrate S, a real figure pattern close to the original figure pattern 10 at the initial design stage can be formed. Thus, by using the figure pattern shape correction apparatus 100 according to the invention of the prior application, the shape of the original figure pattern can be accurately corrected.

  The evaluation point setting unit 110, the feature value extraction unit 120, the bias estimation unit 130, and the pattern correction unit 140 shown in FIG. 1 are all configured by incorporating a predetermined program in a computer. Therefore, the figure pattern shape estimation apparatus 100 'and the figure pattern shape correction apparatus 100 according to the invention of the prior application are actually realized by incorporating a dedicated program into a general-purpose computer.

  FIG. 4 is a flow chart showing a product design / production process using the figure pattern shape correction apparatus 100 shown in FIG. First, in step S1, a product design stage is performed. This product design stage is a process of creating a figure pattern for forming a semiconductor device or the like, and the original figure pattern 10 shown in FIG. 1 is created at this product design stage. In addition, since the apparatus itself which designs products, such as a semiconductor device etc., and produces a figure pattern is already a well-known apparatus, detailed description is abbreviate | omitted here.

  The evaluation point setting step of the next step S2 is a step executed in the evaluation point setting unit 110 shown in FIG. For example, when an original figure pattern 10 as shown in FIG. 2 (a) is given, evaluation points E11, E12, E13 as shown in FIG. 3 (a) are set. The feature quantity extraction step of the subsequent step S3 is a stage executed in the feature quantity extraction unit 120 shown in FIG. 1, and as described above, n kinds of feature quantities x1 to xn are extracted for one evaluation point E. (Detailed extraction procedure is described in § 2). The process bias estimation step of step S4 is a step executed in the bias estimation unit 130 shown in FIG. 1, and as described above, a process is performed using n feature quantities x1 to xn for one evaluation point E. A bias estimate y is obtained (a detailed calculation procedure will be described in § 3).

  Then, the pattern shape correction step of step S5 is a step executed in the pattern correction unit 140 shown in FIG. 1, and as described above, using the process bias estimated value y obtained for each evaluation point E, By correcting the pattern 10, a corrected figure pattern 15 is obtained. Since such correction is not sufficient only once, the process of returning to step S2 is repeated until it is determined that "correction is completed" in step S6. That is, by treating the corrected figure pattern 15 obtained in step S5 as a new original figure pattern 10, the processes of steps S2 to S5 are repeatedly executed.

  As a result of such repeated processing, when it is determined that “correction is completed” in step S6, the process proceeds to step S7, and the lithography process is performed. The determination of "correction completed" is, for example, that "the difference between the position on the original figure pattern and the position on the figure pattern obtained by simulation for a certain percentage of evaluation points E is less than a predetermined reference value" Specific conditions are met. In the lithography process of step S7, an actual process of exposure, development, and etching is performed based on the finally obtained corrected figure pattern, and the actual substrate S is manufactured.

  In the flowchart shown in FIG. 4, the stages of steps S1 to S6 are processes executed on a computer, and the stage of step S7 is a process executed on the actual substrate S.

<1.3 Basic concept of feature quantity extraction in prior invention>
Up to this point, the basic configuration and basic operation of the figure pattern shape estimation device 100 'and figure pattern shape correction device 100 shown in FIG. 1 have been described. In each of these devices, the most characteristic component of the invention of the prior application is the feature value extraction unit 120. The prior application invention has an operation effect of extracting the accurate feature amount from the original figure pattern 10 and performing an accurate simulation to accurately estimate the shape of the actual figure pattern 20 formed on the actual substrate S. However, the component that plays the most important role in obtaining such effects is the feature extraction unit 120. In other words, the important feature of the invention of the prior application lies in the extraction of feature quantities from the original graphic pattern 10 in a very unique manner. Therefore, here, the basic concept of feature quantity extraction in the prior application invention will be described.

  FIG. 5 is a plan view showing a concept of grasping features around the evaluation points E11, E12 and E13 defined on the outline of the rectangular figure pattern 10. As shown in FIG. FIG. 5A shows a state in which the features of the reference circle C1 and the reference circle C2 are extracted at the evaluation point E11 set at the center of the right side of the rectangle. Although the reference circles C1 and C2 are both circles centered on the evaluation point E11, the reference circle C2 is a larger circle than the reference circle C1. Similarly, FIG. 5 (b) shows the evaluation point E12 set at the lower right side of the rectangle, and FIG. 5 (c) shows two reference circles C1, C2 at the evaluation point E13 set at the center of the lower side of the rectangle. The state which extracted the internal feature of C2 is shown.

  Here, for each evaluation point, when the internal feature of the reference circle C1 is compared, the left half of the evaluation point E11 and the evaluation point E12 is inside the figure (hatched area) and the right half is outside the figure (blank area). There is no difference in the internal features of the reference circle C1. On the other hand, the internal feature of the reference circle C1 for the evaluation point E13 is that the upper half is inside the figure (hatched area) and the lower half is outside the figure (blank area), and within the reference circle C1 of the evaluation points E11 and E12. It is a 90 ° rotation of the feature of (1), but there is no difference in the occupancy of the hatched area. On the other hand, when the internal characteristics of the reference circle C2 are compared for each of the evaluation points E11, E12, and E13, it can be seen that the distributions of hatching regions are different from each other and have different characteristics.

  Thus, for each evaluation point E11, E12, E13 on the graphic pattern 10, even if the characteristics of the vicinity thereof are grasped, whether the characteristics of a narrow adjacent region such as the reference circle C1 are extracted or the reference circle C2 is extracted The extracted features differ depending on whether or not the features of the slightly wider neighborhood region are extracted. Therefore, in the case of quantitatively extracting the feature of the neighboring region as a certain feature amount x for a certain evaluation point E, if the range of the neighboring region is changed stepwise, the feature amount can be extracted by more various methods. It can be seen that

  Also, as described above, when the lithography process is performed based on the original figure pattern 10, the actual figure pattern 20 obtained on the substrate has a dimensional error (process bias relative to the original figure pattern 10 due to the proximity effect in the exposure process. ) Will occur. In particular, the proximity effect in electron beam exposure includes various effects such as an effect due to forward scattering having a narrow influence range and an effect due to backscattering having a wide influence range. For example, forward scattering is described as a phenomenon in which electrons of small mass are spread while being scattered by molecules in the resist when the electron beam is irradiated to the layer to be formed including the resist layer or the like, and backscattering is The phenomenon is described as a phenomenon in which electrons scattered and reflected around the surface of a metal substrate or the like under the resist layer are diffused in the resist layer. In addition, a process bias is also generated by the etching process, and the size thereof changes according to the loading phenomenon at the time of etching. Similarly to the proximity effect in the above-described exposure process, this loading phenomenon also results from the combination of various components such as a component having a narrow influence range and a component having a wide influence range.

  After all, the value of the process bias y for one evaluation point E becomes a value determined by fusing phenomena with various scales. Therefore, extracting various feature amounts from the feature amount for the narrow range to the feature amount for the wide range is different for the various phenomena in each process affecting the process bias, each having a different range of influence. Important for accurate simulation. Therefore, in the prior application invention, for one evaluation point E, feature amounts for various regions around it are extracted from near to far. As described above, in order to extract a plurality of feature amounts for one evaluation point E, the prior invention adopts a method of creating an "image pyramid" composed of a plurality of hierarchical images having different sizes. The image pyramid contains information in which various phenomena having different influence ranges are multiplexed.

  FIG. 6 is a diagram showing an outline of processing executed in the feature quantity extraction unit 120 and the bias estimation unit 130 shown in FIG. The original image Q1 shown at the top of the figure is an image created by the original image creation unit 121 shown in FIG. 1 and is an image corresponding to the given source graphic pattern 10. As described above, the original figure pattern 10 is data created by a design device of a semiconductor device or the like, and becomes data indicating a figure as shown in FIG. 2A, but usually indicates an outline of the figure. It is given as vector data (data indicating the connection between the coordinate value of each vertex and each vertex).

  The original image creation unit 121 executes a process of creating an original image Q1 composed of a set of pixels each having a predetermined pixel value based on the given data of the original graphic pattern 10. For example, if a pixel having a pixel value of 1 is arranged inside the figure constituting the original figure pattern 10 and a pixel having a pixel value of 0 is arranged outside, the original image Q1 consisting of a collection of many pixels U Can be created. The original image Q1 shown in FIG. 6 is an image composed of such a set of pixels U, and has rectangular figures included in the original figure pattern 10 as image information, as shown by the broken lines. Further, the evaluation point setting unit 110 sets an evaluation point E on the outline of this figure. Although only one evaluation point E is depicted in FIG. 6 for the sake of convenience, in actuality, a large number of evaluation points are set along the outline of the figure.

  The image pyramid creation unit 122 shown in FIG. 1 creates an image pyramid PP based on the original image Q1. The image pyramid PP is composed of a plurality of hierarchical images each having a different size. The figure shows an image pyramid PP composed of a plurality of (nn2) hierarchical images P1 to Pn. Although a specific procedure for creating hierarchical images P1 to Pn from the original image Q1 will be described in Section 2, basically, the hierarchical images P1 to Pn are created by reduction processing to reduce the number of pixels. For example, in the illustrated example, the hierarchical image P1 is an image of the same size as the original image Q1 (image having the same number of vertical and horizontal pixels), whereas the hierarchical image P2 is a reduced-size image of small size The hierarchical image P3 is a smaller size image obtained by further reducing the hierarchical image P2.

  As described above, in the prior invention, hierarchical images P1 to Pn are created such that the image size gradually decreases based on the original image Q1. As described above, a state in which a plurality of hierarchical images having different sizes are vertically arranged is in the form of a pyramid as shown in the figure, so in the prior application, an assembly of the plurality of hierarchical images P1 to Pn is referred to as an image pyramid PP. I'm calling. Since all of the hierarchical images P1 to Pn are created based on the original image Q1, the hierarchical images P1 to Pn have information of the original figure pattern 10, and the position of the evaluation point E can be defined for each of them. In the figure, rectangular figures are drawn on each of the hierarchical images P1 to Pn, and evaluation points E are arranged on the outlines thereof.

  The feature quantity calculation unit 123 illustrated in FIG. 1 performs a process of calculating feature quantities based on the pixel values of the pixels in the vicinity of the evaluation point for each hierarchical image constituting the image pyramid. FIG. 6 shows a state in which the feature quantities x1 to xn of the evaluation point E are extracted from the n hierarchical images P1 to Pn constituting the image pyramid PP. The illustrated feature quantities x1 to xn are values indicating features around the same evaluation point E, but the feature quantity x1 is a pixel value of a pixel near the evaluation point E on the first hierarchical image P1. The feature amount x2 is a value calculated based on the pixel values of pixels in the vicinity of the evaluation point E on the second hierarchical image P2, and the feature amount x3 is a third value. This is a value calculated based on the pixel values of the pixels in the vicinity of the evaluation point E on the hierarchical image P3. A specific calculation procedure of each of the feature amounts x1 to xn will be described in §2.

  Although only one evaluation point E is depicted in FIG. 6 for the sake of convenience, n feature amounts x1 to xn are actually calculated for each of a large number of evaluation points. Although each of the feature quantities x1 to xn is a predetermined scalar value, n feature quantities x1 to xn are obtained for each evaluation point E, respectively. Therefore, if the n feature quantities x1 to xn are grasped as an n-dimensional vector, the feature quantity extraction unit 120 performs processing for extracting a feature quantity consisting of an n-dimensional vector for one evaluation point E. .

  The feature quantity (n-dimensional vector) for each evaluation point extracted in this manner is input by the feature quantity input unit 131 in the bias estimation unit 130 and is given to the estimation operation unit 132. In the case of the embodiment shown here, the estimation operation unit 132 is constituted by a neural network, and based on the learning information L obtained by the learning step performed in advance, the feature quantities x1 to xn given as n-dimensional vectors In response, calculation is performed to calculate the estimated value y (scalar value) of the process bias for the evaluation point E. Specific calculation procedures will be described in §3.

  Thus, according to the prior application invention, accurate feature quantities are extracted for every original figure pattern 10 even if no physical / experimental simulation model is built for the actual lithography process. be able to. In addition, since it is not necessary to construct physical and experimental simulation models, it is not necessary to consider various set values of material physical properties and process conditions in the learning phase of the neural network described later in 3.2 3.2.

  In the above, an example of using the shape estimation device 100 'of the figure pattern and the shape correction device 100 according to the invention of the prior application in the lithography process for manufacturing a semiconductor device has been described. The present invention is not limited to the application to the manufacturing process of the present invention, and can be applied to the manufacturing process of various products including a lithography process. For example, the invention of the prior application or the present invention can be used in the manufacturing process of various products including lithography processes such as NIL (Nano Imprint Lithography) and EUV (Extreme UltraViolet Lithography). In particular, in NIL, correction may be performed on the original figure pattern so that the actual figure pattern on the master template produced from the original figure pattern through exposure lithography matches the original figure pattern. Alternatively, correction may be performed on the original figure pattern so that the actual figure pattern on the replica template produced through imprinting from the master template matches the original figure pattern. In addition, the prior application invention and the invention of the present application are all product fields including lithography processes, such as Micro Electro Mechanical Systems (MEMS), Large-size Photomask (LSPM), lead frames, metal masks, metal mesh sensors, color filters, etc. Applicable to

<<< 2. 2. Details of feature extraction unit >>>
Subsequently, the detailed processing operation of the feature quantity extraction unit 120 will be described. As shown in FIG. 1, the feature quantity extraction unit 120 has an original image creation unit 121, an image pyramid creation unit 122, and a feature quantity calculation unit 123, and the feature quantity extraction process of step S3 in the flowchart of FIG. It has a function to execute. This feature quantity extraction process is actually performed according to each procedure shown in FIG. Here, steps S31 and S32 are procedures executed by the original image creation unit 121, steps S33 to S36 are procedures executed by the image pyramid creation unit 122, and step S37 is a feature quantity calculation unit 123. Is a procedure performed by Hereafter, the procedure performed by each part is demonstrated in detail, giving a specific example.

<2.1 Processing Procedure by Original Image Creation Unit 121>
The original image creation unit 121 has a function of creating an original image composed of a set of pixels each having a predetermined pixel value based on the given source graphic pattern 10, and performs steps S31 and S32 of the flowchart of FIG. Run. First, in step S31, a process of inputting the original graphic pattern 10 is performed, and in the subsequent step S32, an original image creation process is performed.

  In § 1, for convenience of explanation, a simple pattern consisting of only one rectangular figure as shown in FIG. 2A is shown as an example of the original figure pattern 10 input in step S 31. Here, in order to explain in more detail, let us consider the case in which the original figure pattern 10 including five figures F1 to F5 (rectangles) as shown in FIG. 8 is given. As described above, the original figure pattern 10 given to the figure pattern shape correction apparatus 100 is usually vector data indicating the outline of the figure. Therefore, FIG. 8 shows hatching in each of the figures F1 to F5. Actually, in the original figure pattern 10, the coordinate values of the vertices of the five rectangles F1 to F5 and the respective vertices are shown. It is given as vector data indicating the connection relation of

  The original image creation process of step S32 is a process of creating data (raster data) of the original image Q1 composed of a group of pixels based on the original graphic pattern 10 given as such vector data. Specifically, the original image creation unit 121 defines a mesh consisting of a two-dimensional array of pixels U, superimposes the original figure pattern 10 on this mesh, and configures the position of each pixel U and the original figure pattern 10 A process of determining the pixel value of each pixel U is performed based on the relationship with the positions of the outlines of the figures F1 to F5.

  FIG. 9 is a plan view showing a state in which processing of superposing the original figure pattern 10 on the mesh formed by the two-dimensional array of the pixels U in the original image creation unit 121 is performed. In this example, a mesh is defined in which pixels U having pixel dimensions u in the vertical and horizontal directions are two-dimensionally arrayed, and a large number of pixels U are arranged at the pitch u in the vertical and horizontal directions. The pixel size u is set to an appropriate value that can represent the shapes of the figures F1 to F5 with a sufficient resolution. As the pixel size u is set smaller, the resolution of shape representation is improved, but the processing load on the later becomes heavier. In general, since the original figure pattern 10 used to manufacture a semiconductor device has a very fine pattern, it is preferable to set a value of about 5 to 20 nm, for example, as the pixel size u.

  Thus, when the two-dimensional array of the pixels U is defined, pixel values are defined for the individual pixels U based on the relationship with the positions of the outlines of the figures F1 to F5. There are several ways to define pixel values.

  The most basic definition method recognizes the internal area and external area of each figure F1 to F5 based on the original figure pattern 10, and sets the occupancy rate of the internal area in each pixel U as the pixel value of the pixel It is a method. In FIG. 8, the hatched area is the inner area of each of the figures F1 to F5, and the white area is the outer area. Therefore, when this method is adopted, the pixel value of each pixel is defined as the occupancy rate (0 to 1) of the internal region (hatched region) in the pixel in the superimposed state as shown in FIG. An image in which pixel values are defined in this manner is generally called an "area density map".

  FIG. 10 is a view showing an area density map M1 created on the basis of “original graphic pattern 10 + two-dimensional array of pixels” shown in FIG. Here, each cell is each pixel defined in FIG. 9, and the numbers in the cell are pixel values defined for each pixel. The blank cell is a pixel having a pixel value 0 (illustration of the pixel value 0 is omitted). In the area density map M1, for example, a pixel having a pixel value of 1.0 is a pixel having 100% occupancy of the hatched area in FIG. 9, and a pixel having a pixel value of 0.5 has a hatched area in FIG. The occupancy rate of the pixel is 50%. Basically, the area density map M1 is a binary image in which the inside of the figure is represented by pixel value 1 and the outside of the figure is represented by pixel value 0. Since a value indicating the ratio is given as a pixel value, the whole is a monochrome gradation image.

  As another method of defining pixel values, a method of recognizing the outline of each figure F1 to F5 based on the original figure pattern 10, and setting the length of the outline existing in each pixel U as the pixel value of the pixel Can also be taken. When this method is adopted, in the state of being superimposed as shown in FIG. 9, the pixel value of each pixel is defined as the sum of the lengths of the outlines present in the pixel. An image in which pixel values are defined in this manner is generally called an "edge length density map". As a unit of the length of the outline, for example, a unit having a pixel size u of 1 may be used.

  FIG. 11 is a view showing an edge length density map M2 created on the basis of “original graphic pattern 10 + two-dimensional array of pixels” shown in FIG. Here again, each cell is each pixel defined in FIG. 9, and the numbers in the cell are defined as the sum of the lengths of the outlines present in each pixel (the length with the pixel size u being 1). Pixel value. The blank cell is a pixel having a pixel value 0 (illustration of the pixel value 0 is omitted). In the edge length density map M2, for example, a pixel having a pixel value of 1.0 is a pixel having a length u of an outline existing in the pixel in FIG. Since this edge length density map M2 is basically a monochrome gradation image showing the density distribution of outlines, it becomes an image having a property considerably different from the area density map M1 described above. However, it is a very useful image in extracting feature quantities for evaluation points E defined on outlines.

  Although the method of defining the pixel value of each pixel has been described above based on the original figure pattern 10 shown in FIG. In addition to the geometric information of the outlines, further information on the dose for each figure may be added. FIG. 12 is a plan view showing an original figure pattern 10 including such dose information. Similar to the original figure pattern 10 shown in FIG. 8, the original figure pattern 10 shown in FIG. 12 includes the information of the outlines of the figures F1 to F5, but in addition to that, For each of F5, it contains information that defines the dose amount.

  Specifically, in the case of the illustrated example, a dose of 100% is defined for figures F1 to F3, a dose of 50% is defined for figure F4, and a dose of 10% is defined for figure F5. . These dose amounts indicate the intensity of light or electron beam to be irradiated (including the case of controlling the total energy amount by the number of exposures) in the exposure step of the lithography process. In the case of the illustrated example, light or electron beam is irradiated with an intensity of 100% when exposing the inner area of the figures F1 to F3, but 50% when exposing the inner area of the figure F4, When exposing the inner area, light or electron beam is irradiated at an intensity of 10%. As described above, if the dose amount is controlled for each figure at the time of the exposure process, the size of the actual figure pattern 20 formed on the actual substrate S can be adjusted more finely.

  In the case where such dose control is performed in the exposure process of the lithography process, it is necessary to consider the dose also in the simulation. In such a case, when determining the pixel value of the original image, it is necessary to adopt a method in which the dose amount is taken into consideration. That is, when an original figure pattern 10 including information on an outline indicating the boundary between the inside and the outside of the figure and information on a dose amount of each figure in the lithography process is given, the original figure pattern 10 is given. Based on the internal region and the external region of each figure, and further recognize the dose amount for each figure, and for each figure existing in each pixel, A method may be employed in which a product is obtained and the sum of the products is used as the pixel value of the pixel.

  When this method is adopted, the pixel value of each pixel in the superimposed state as shown in FIG. 9 is the occupancy rate (0 to 1) of the internal area (hatched area) of the specific figure in the pixel and the specific It is defined as the sum of the product of the dose amount of the figure. An image in which pixel values are defined in this manner is generally called a "dose density map". This dose density map is also a monochrome gradation image as a whole.

  FIG. 13 is a view showing a dose density map M3 created based on the original figure pattern 10 with a dose amount shown in FIG. Here, each cell is each pixel defined in FIG. 9, and the numbers in the cell are pixel values defined for each pixel. The blank cell is a pixel having a pixel value 0 (illustration of the pixel value 0 is omitted). When this dose density map M3 is compared with the area density map M1 shown in FIG. 10, there is no change in the pixel value for the pixel where the figures F1 to F3 given the dose amount 100% are arranged, but the dose amount 50% It can be seen that the pixel value for the pixel where the figure F4 given by and the figure F5 given a dose amount of 10% are arranged is reduced by an amount according to the dose amount. This shows a phenomenon that light or electron beam whose intensity is reduced is irradiated to the inside of the figures F4 and F5 in an actual exposure process.

  In the above, based on the original figure pattern 10, an example of creating three kinds of images (aggregation of pixels U having predetermined pixel values) of area density map M1, edge length density map M2, and dose density map M3 has been shown. . In the original image creation processing of step S32 in the flowchart of FIG. 7, any one of these three types of images may be created and used as the original image. FIG. 7 shows an example in which the area density map M1 is the original image Q1 as an example. Of course, it is also possible to create the original image Q1 by methods other than the three methods described above.

  The original image Q1 created in step S32 is used as a first preparation image Q1 in the filter process of step S33. The filter process of step S33 is a process of creating an k-th hierarchical image Pk by applying an image processing filter to the k-th preparation image Qk, as described later. Therefore, the procedure of step S32 can be said to be processing of setting the parameter k (k = 1, 2, 3,...) To the initial value k = 1 and creating the first preparation image Q1. . The first preparation image Q1 is an original image created first based on the original graphic pattern 10, and is a reference image used first in the image pyramid creating process described later.

<2.2 Processing Procedure by Image Pyramid Creation Unit 122>
Next, the procedure of the process of creating an image pyramid by the image pyramid creation unit 122 will be described. The image pyramid creation unit 122 has a function of performing reduction processing to reduce the number of pixels based on the original image Q1 (for example, the area density map M1 shown in FIG. 10) created in step S32 of FIG. An image pyramid creation process is performed to create an image pyramid composed of a plurality of hierarchical images having different sizes. In the case of the embodiment described herein, this image pyramid creation process is performed by the procedure shown in steps S33 to S36 of the flowchart of FIG.

  Here, in the step S31, the original graphic pattern 10 as shown in FIG. 8 is input, and in the step S32, a specific image is taken as an example where the area density map M1 shown in FIG. 10 is created as the original image Q1. The procedure of the pyramid creation process will be described.

  First, in step S33, an image processing filter is applied to the k-th preparation image Qk to perform filter processing for creating the k-th hierarchical image Pk. Specifically, this filtering process is performed as, for example, a convolution operation using a Gaussian filter as an image processing filter. FIG. 14 is a plan view showing a procedure for creating the k-th hierarchical image Pk by applying a filtering process using a Gaussian filter GF33 to the preparation image Qk. The k-th preparation image Qk shown in FIG. 14 is actually the same as the area density map M1 shown in FIG.

  That is, the area density map M1 shown in FIG. 10 is shown as an 8 × 8 pixel array for convenience, and the description of the pixel value 0 is omitted, whereas the preparation image Qk shown in FIG. Although shown as an array of 10 pixels and the pixel value 0 is also described, both are substantially the same image. In short, in FIG. 14, in order to execute the filter processing, pixels having a pixel value 0 are arranged around the area density map M1 consisting of the 8 × 8 pixel array shown in FIG. It is only At this stage, the parameter k has an initial value of 1, and the preparation image Qk shown in FIG. 14 is the first preparation image Q1. As described above, the first preparation image Q1 is nothing but the original image created in the original image creation process in step S32.

  In the filter process shown in FIG. 14, a convolution operation using a Gaussian filter GF33 is performed. The Gaussian filter GF33 is a 3 × 3 pixel array as shown in the drawing, and the k-th is performed by superimposing the Gaussian filter GF33 on a predetermined position of the k-th preparation image Qk and performing product-sum operation processing. Hierarchical images Pk (filtered images) are obtained. FIG. 15 is a plan view showing the k-th hierarchical image Pk obtained by the filter processing shown in FIG. The k-th hierarchical image Pk has a 10 × 10 pixel array similarly to the k-th preparatory image Qk, and the pixel values of the individual pixels are obtained by product-sum operation using the Gaussian filter GF33. It becomes a value.

  For example, in the k-th hierarchical image Pk shown in FIG. 15, focusing on the pixels (pixels in the fourth row and the third column) surrounded by a thick frame, a pixel value of 0.375 is given to the target pixel . The pixel value is obtained by superposing the illustrated Gaussian filter GF33 on the 3 × 3 pixel array (9 pixels centered on the fourth row and third column pixels) surrounded by a bold frame in FIG. The product of the pixel values of the pixels overlapped at the same position is determined as the sum of the nine products. Specifically, the pixel value 0.375 of the pixel of interest is “(1/16 × 0) + (2/16 × 0.25) + (1/16 × 0.5) + (2/16 × 0) ) + (4/16 x 0.5) + (2/16 x 1.0) + (1/16 x 0) + (2/16 x 0.25) + (1/16 x 0.5) " The product-sum operation value is obtained as Such filter processing by product-sum operation is a process generally known as a convolution operation process on an image, and thus the detailed description is omitted here.

  In the filter processing shown in FIG. 14, the convolution operation is performed using a Gaussian filter GF33 consisting of a 3 × 3 pixel array as shown in FIG. 16A as an image processing filter. The convolution operation may be performed using a Laplacian filter LF33 having a 3 × 3 pixel arrangement as shown in b). In general, it is known that filtering using a Gaussian filter has an effect of blurring the outline of an image, and filtering using a Laplacian filter has an effect of enhancing the outline of an image. No matter which filter process is adopted, the k-th layered image Pk having slightly different features can be obtained with respect to the k-th preparation image Qk, and therefore, a plurality of layered images having different features are provided. It is effective in creating an image pyramid.

  Of course, the image processing filter used for the filter processing in step S33 is not limited to the Gaussian filter GF33 shown in FIG. 16 (a) or the Laplacian filter LF33 shown in FIG. 16 (b). It is possible to use processing filters. Also, the size of the image processing filter to be used is not limited to the 3 × 3 pixel array, and it is possible to use an image processing filter of any size. For example, a Gaussian filter GF55 consisting of a 5 × 5 pixel array as shown in FIG. 17 (a) or a Laplacian filter LF55 consisting of a 5 × 5 pixel array as shown in FIG. 17 (b) may be used. .

  Thus, when the filtering process of step S33 in the procedure of FIG. 7 is completed, it is determined in step S34 whether or not the parameter k has reached a predetermined set value n. If k <n, reduction of step S35 is performed. Processing is performed. This reduction process is a process of creating an image having a smaller number of pixels than the target image based on a predetermined target image. In the case of the embodiment shown here, the (k + 1) -th preparation image Q (k + 1) is created by executing the reduction process on the k-th hierarchical image Pk created in the filter process of step S33. Processing is performed. Therefore, the preparation image Q (k + 1) is an image smaller in size than the hierarchical image Pk (an image in which the number of vertical and horizontal pixels in the pixel array is small).

  The reduction process for such an image is also referred to as "pooling process", and for example, "average pooling process" can be adopted as the reduction process performed in step S35. FIG. 18 is a plan view showing a procedure for creating the (k + 1) th preparation image Q (k + 1) as a reduced image by performing average pooling processing on the k-th hierarchical image Pk. . Specifically, the average pooling process (reduction process) is applied to the hierarchical image Pk consisting of the 4 × 4 pixel array shown in FIG. 18A to obtain 2 × 2 image shown in FIG. 18B. A preparation image Q (k + 1) composed of a pixel array is created as a reduced image.

  The average pooling process shown in FIG. 18 is a process of converting (reducing) four pixels consisting of a 2 × 2 pixel array into one pixel, and calculates the average value of the pixel values of the original four pixels, By using the pixel value of one pixel after conversion, a reduced image is created. For example, four pixels (pixels within a bold frame) consisting of a 2 × 2 pixel array disposed at the upper left of the hierarchical image Pk shown in FIG. 18A are the preparation image Q (FIG. 18B). On k + 1), conversion (reduction) is performed to one pixel indicated by a thick frame. The pixel value 0.5 of the thick frame pixel after this conversion (reduction) is the average value of the pixel values of the original four pixels.

  On the other hand, FIG. 19 is a plan view showing a procedure for creating the (k + 1) th preparation image Q (k + 1) as a reduced image by performing max-pooling processing on the k-th hierarchical image Pk. It is. Specifically, by performing max pooling processing (reduction processing) on the hierarchical image Pk consisting of the 4 × 4 pixel array shown in FIG. 19A, the 2 × 2 image shown in FIG. 19B is obtained. A preparation image Q (k + 1) composed of a pixel array is created as a reduced image.

  The max pooling process shown in FIG. 19 is a process to convert (reduce) four pixels consisting of a 2 × 2 pixel array into one pixel, similarly to the average pooling process shown in FIG. A reduced image is created by setting the maximum value of the pixel values of the four pixels in the above as the pixel value of one pixel after conversion. For example, four pixels (pixels within a bold frame) consisting of a 2 × 2 pixel array arranged at the upper left of the hierarchical image Pk shown in FIG. 19A are the preparation image Q (FIG. 19B). On k + 1), conversion (reduction) is performed to one pixel indicated by a thick frame. The pixel value 1.0 of the thick frame pixel after this conversion (reduction) is the maximum value of the pixel values of the original four pixels.

  Although each pooling process shown in FIGS. 18 and 19 is a reduction process for converting four pixels consisting of a 2 × 2 pixel array into a single pixel, it is of course made up of a 3 × 3 pixel array It is also possible to perform a reduction process to convert nine pixels into a single pixel, or to perform a reduction process to convert six pixels of a 3 × 2 pixel array to a single pixel. It is.

  After all, as the reduction process of step S35, the image pyramid creating unit 122 replaces the plurality of m adjacent pixels with a single pixel whose pixel value is the average value of the pixel values of the plurality of m adjacent pixels. It is possible to create a reduced image by performing pooling processing, or to replace a plurality of m adjacent pixels with a single pixel whose pixel value is the maximum value of the pixel values of the plurality m of adjacent pixels. A reduced image can also be created by performing pooling processing. Of course, other reduction processes can also be performed as the reduction process of step S35. In short, by performing conversion to reduce the number of pixels with respect to the hierarchical image Pk, a process capable of creating a small-sized preparatory image Q (k + 1), in other words, “the number of pixels has decreased Any reduction process may be performed in step S35 as long as it is a process of creating a “reduced image”.

  Thus, when the reduction process of step S35 is completed, the parameter k is increased by 1 in step S36, and the filter process of step S33 is performed again. As a result, as described above, the first hierarchical image P1 is created by performing the filtering process in step S33 on the first preparation image Q1 (original image) with k = 1, and then the step S35. Then, the second preparation image Q2 is created by performing reduction processing on this hierarchical image P1, and is updated to k = 2 in step S36, and again in step S33, a filter for the second preparation image Q2 By performing the processing, the second hierarchical image P2 is created.

  Such a repetition procedure is repeatedly performed until it is determined that k = n in step S34. Here, as the value of n, an appropriate value may be set in advance as the number of layers of the image pyramid (that is, the total number of layer images constituting the image pyramid). As the value of n is set larger, the number n of feature quantities extracted for one evaluation point E increases, so that more accurate simulation is possible, but the computational load increases. Further, as the reduction process of step S35 is repeated, the size of the image decreases more and more, so if the value of n is set too large, the reduction process of step S35 can not be performed. Therefore, in practice, the value of n may be set appropriately in consideration of the size of the original image Q1 and the computational burden.

  Thus, when it is determined that k = n in step S34, the processing by the image pyramid creating unit 122 is completed. At this point of time, as shown in FIG. 6, an image pyramid PP composed of a plurality n of hierarchical images P1 to Pn is created. Therefore, the process proceeds from step S34 to step S37, and the feature amount calculation process is executed.

  FIG. 20 is a plan view showing a procedure (procedure of steps S33 to S36 in FIG. 7) of creating the image pyramid PP composed of n kinds of hierarchical images P1 to Pn in the image pyramid creating unit 122. The first preparation image Q1 shown at the upper left of the figure is an original image generated with k = 1 in the original image generation processing of step S32, and for each pixel, for example, an area density map M1 shown in FIG. Pixel values such as are defined. As described above, in step S33, filter processing is performed on the first preparation image Q1. Specifically, for example, a first hierarchical image P1 as shown in the upper right of FIG. 20 is created by a convolution operation using a Gaussian filter GF33 consisting of a 3 × 3 pixel array. The size of the first hierarchical image P1 is the same as the size of the first preparation image Q1.

  Subsequently, in the reduction process of step S35, the reduction process (for example, average pooling process) is performed on the first hierarchical image P1, and a second preparation image Q2 shown in the middle left of FIG. 20 is created. The size of the second preparation image Q2 is smaller than the size of the first hierarchical image P1. Subsequently, in step S36, the value of the parameter k is updated to 2, and the filtering process of step S33 is performed again. That is, the second hierarchical image P2 as shown in the middle right of FIG. 20 is created by the convolution operation using the Gaussian filter GF33 consisting of a 3 × 3 pixel array. The size of the second hierarchical image P2 is the same as the size of the second preparation image Q2.

  Then, the reduction process of step S35 is executed again. That is, reduction processing (for example, average pooling processing) is performed on the second hierarchical image P2, and a third preparation image Q3 shown in the lower left of FIG. 20 is created. The size of the third preparation image Q3 is smaller than the size of the second hierarchical image P2. Subsequently, in step S36, the value of the parameter k is updated to 3, and the filtering process of step S33 is performed again. That is, the third hierarchical image P3 as shown in the lower right of FIG. 20 is created by the convolution operation using the Gaussian filter GF33 consisting of a 3 × 3 pixel array. The size of the third hierarchical image P3 is the same as the size of the third preparation image Q3.

  Such processing is repeatedly performed until the parameter k = n, and finally, the nth preparation image Qn and the nth hierarchy image are obtained. Thus, the image pyramid PP is configured by n hierarchical images different in size from the first hierarchical image P1 to the n-th hierarchical image Pn.

  After all, in the case of the embodiment shown in the procedure of FIG. 7, the image pyramid creation unit 122 has a function of performing filter processing using a predetermined image processing filter on the original image Q1 or the reduced image Q (k + 1). By alternately executing the filtering process and the reduction process, an image pyramid PP composed of a plurality of hierarchical images P1 to Pn is created.

  More specifically, the image pyramid creation unit 122 sets the original image created by the original image creation unit 121 as the first preparation image Q1, and filters the kth preparation image Qk (where k is a natural number). Let the resulting image be the kth hierarchical image Pk, and let the image obtained by the reduction process on the kth hierarchical image Pk be the (k + 1) th preparatory image Q (k + 1) until the nth hierarchical image Pn is obtained By alternately executing the process and the reduction process, an image pyramid PP made up of a plurality of n-layer images including the first layer image P1 to the n-th layer image Pn is created.

  Since the image pyramid PP in the prior application invention may be formed of a plurality of hierarchical images having different sizes, the reduction process of step S35 is an essential process in the procedure shown in the flowchart of FIG. However, the filtering process of step S33 is not necessarily a necessary process. However, if filter processing is performed, the pixel value of each pixel can be influenced by the pixel value of the surrounding pixel. For this reason, by adding the filter processing, it is possible to create a plurality of hierarchical images rich in variations, and it becomes possible to extract feature quantities containing more diverse information, and as a result, it is more accurate. Simulation is possible. Therefore, practically, it is preferable to alternately execute the reduction process and the filtering process as shown in the flowchart of FIG. 7.

<2.3 Processing Procedure by Feature Amount Calculation Unit 123>
Next, the procedure of feature amount calculation processing by the feature amount calculation unit 123 will be described. As shown in step S37 of FIG. 7, the feature amount calculation unit 123 performs processing of calculating feature amounts x1 to xn for each evaluation point E based on the hierarchical images P1 to Pn constituting the image pyramid PP. Here, the procedure of the process of calculating the feature quantities x1 to xn will be specifically described.

  FIG. 21 is a plan view showing a procedure for calculating the feature amounts x1 to xn for a specific evaluation point E from the layer images P1 to Pn in the feature amount calculation unit 123. Specifically, on the left side of FIGS. 21A to 21C, the original graphic pattern 10 is formed in each pixel array of the first hierarchical image P1, the second hierarchical image P2, and the third hierarchical image P3. 21 (a) to (c) show specific evaluation points based on the hierarchical images P1, P2, and P3 on the right side. The principle of calculating feature quantities x1, x2, x3 for E is shown.

  Each hierarchy image P1, P2, P3 shown in FIG. 21 is a part of the image which comprises each hierarchy of image pyramid PP. Actually, n kinds of hierarchy images from P1 to Pn are prepared, and n sets of feature quantities x1 to xn are extracted. However, in FIG. It is shown that three sets of feature quantities x1, x2, x3 are extracted from P2, P3.

  Here, the first hierarchical image P1 is an image obtained by applying the filtering process to the original image Q1 (first preparation image), and in the illustrated example, has a 16 × 16 pixel array. doing. On the other hand, the second hierarchical image P2 is an image obtained by performing the reduction process and the filter process on the first hierarchical image P1, and in the illustrated example, the 8 × 8 pixel array is Have. The third hierarchical image P3 is an image obtained by performing the reduction process and the filter process on the second hierarchical image P2, and in the illustrated example, has a 4 × 4 pixel array. There is.

  In FIG. 21, the hierarchical images P1, P2 and P3 are drawn so that their outlines become squares of the same size, and thus they are all images of the same size. The size is gradually reduced to × 16, 8 × 8, 4 × 4, and the size of the image is gradually reduced. However, in FIG. 21, since the outer frames of the hierarchical images P1, P2, and P3 are drawn as squares having the same size, the size of the pixels is gradually increased. In other words, the resolution of the image decreases in the order of the hierarchical images P1, P2, and P3 and gradually becomes coarse.

  As described above, the rectangles forming the original figure pattern 10 are drawn in a thick frame in the figure. Each hierarchical image P1, P2, P3 is a raster image consisting of a group of pixels, so a rectangular outline drawn in bold in the figure is actually included as information of the outline itself. It does not mean that it is included as information of the pixel value of each pixel. However, in FIG. 21, for convenience of explanation, the positions of the rectangles on the hierarchical images P1, P2, and P3 are indicated by thick lines. Here, a process of extracting the feature amounts x1 to xn will be described for the specific evaluation point E defined on the contour line of the rectangle.

  As can be seen from FIGS. 21 (a) to 21 (c), the rectangles shown by thick frames are arranged at the same relative position with respect to each hierarchical image P1, P2 and P3, and the specific evaluation points E are also the same relative. It is placed in position. In the case of the embodiment shown here, the feature amount for one evaluation point E is calculated based on the pixel values of the pixels in the vicinity thereof.

  First, as shown in FIG. 21A, the feature amount x1 for the evaluation point E is extracted based on the first hierarchical image P1. Specifically, as shown on the right side of FIG. 21A, the feature quantity calculation unit 123 determines that four pixels located in the vicinity of the evaluation point E from the pixels forming the first hierarchical image P1 (see FIG. Is extracted as a pixel of interest, and the feature amount x1 is calculated by calculation using the pixel values of these four pixels of interest. Similarly, as shown on the right side of FIG. 21B, attention is focused on four pixels (pixels hatched in the figure) located in the vicinity of the evaluation point E from the pixels constituting the second hierarchical image P2. A feature amount x2 is calculated by extraction as a pixel and calculation using pixel values of these four target pixels. In addition, as shown on the right side of FIG. 21C, from the pixels forming the third hierarchical image P3, four pixels (pixels hatched in the figure) located in the vicinity of the evaluation point E are noted pixels Are extracted, and the feature amount x3 is calculated by calculation using the pixel values of these four target pixels.

  If such processing is performed also for the fourth layer image P4 to the n-th layer image Pn, n sets of feature amounts x1 to xn can be extracted for a specific evaluation point E. These n sets of feature quantities x1 to xn are parameters indicating features around the same evaluation point E on the original figure pattern 10, but the ranges affected by the original figure pattern 10 are different from each other . For example, the feature quantity x1 extracted from the first hierarchical image P1 has a value indicating the feature in the narrow area hatched in the right side of FIG. 21A, but the second hierarchical image P2 The feature amount x2 extracted from the image becomes a value indicating the feature in a wider area hatched in the right side of FIG. 21B, and the feature amount x3 extracted from the third hierarchical image P3 is The figure on the right side of FIG. 21 (c) is a value indicating the feature in a wider area hatched.

  As described above, the value of the process bias y for one evaluation point E is a value determined by fusing phenomena having various scale feelings such as forward scattering and back scattering. Therefore, if various feature quantities x1 to xn are extracted as feature quantities for the same evaluation point E from feature quantities x1 for a very narrow range around it to feature quantities xn for a wider range, the influence range Can perform accurate simulations considering various phenomena that differ from one another. FIG. 21 shows a process of extracting n sets of feature amounts x1 to xn for one evaluation point E. However, in practice, each of a large number of evaluation points defined on the original figure pattern 10 is shown. , N sets of feature quantities x1 to xn are extracted by the same procedure.

  As a method of calculating the feature amount x based on the pixel value of the target pixel in the vicinity of the evaluation point E, a simple method of using the simple average of the pixel values of the target pixel as the feature amount x can be adopted. For example, as shown in FIG. 21A, in order to extract the feature amount x1 for the evaluation point E from the first layer image P1, the pixel of interest (shown in the vicinity of the evaluation point E) is hatched in the figure. The simple average of pixel values of certain four pixels) may be set as the feature amount x1. However, in order to calculate a more accurate feature quantity, it is preferable to obtain a weighted average in consideration of a weight according to the distance between the evaluation point E and each focused pixel, and to set the value of this weighted average as the feature quantity x1. .

  FIG. 22 is a diagram showing a specific calculation method (a calculation method using a weighted average value as a feature quantity) used in the feature quantity calculation procedure shown in FIG. Here, an example is shown in which four pixels A, B, C, and D are selected as target pixels located in the vicinity of a specific evaluation point E. Specifically, on the hierarchical image P to be processed, the target pixels A, B, C, and D can be determined by performing processing of selecting a total of four pixels in order of closeness to the evaluation point E. Therefore, with respect to the pixel values of the four pixels of interest A, B, C, and D, an operation may be performed in which a weighted average taking into consideration the weight according to the distance between the evaluation point E and each pixel is used as the feature amount x.

  In FIG. 22 (a), x marks are displayed at the center points of the noted pixels A, B, C, D, and broken lines connecting these x marks are drawn. The pixel size of each of the target pixels A, B, C, and D is u in both the vertical and horizontal directions, and the broken line is a dividing line dividing the pixel having the pixel size u into halves. In the case of the embodiment shown here, the distance between the evaluation point E and the central point of each of the target pixels A, B, C, D as the distance between the evaluation point E and each of the target pixels A, B, C, D The vertical distance is adopted. Specifically, in the case of the example shown in FIG. 22 (a), the horizontal distance a and the vertical distance c are for the pixel of interest A, and the horizontal distance b and the vertical distance c for the pixel of interest B. Thus, for the pixel of interest C, the horizontal distance a and the vertical distance d are obtained, and for the pixel of interest D, the horizontal distance b and the vertical distance d.

In this case, if the feature value x of the evaluation point E represents the pixel value of each focused pixel A, B, C, D by the same symbols A, B, C, D, and the pixel size u (pixel pitch) is 1. As shown in FIG. 22 (b),
G = (A · b + B · a) / 2
H = (C · b + D · a) / 2
x = (G · d + H · c) / 2
It can be obtained by the following operation.

  Of course, the method of calculating the feature quantity x from the pixel values of the four pixels of interest A, B, C, and D is not limited to the method illustrated in FIG. If the feature amount x reflecting the pixel value can be calculated, it is possible to adopt various other calculation methods. In the calculation methods illustrated in FIGS. 21 and 22, four pixels located in the vicinity of the evaluation point E are selected as the target pixels, but the number of target pixels used to calculate the feature amount x is It is not limited to four. For example, nine pixels forming a 3 × 3 pixel array located in the vicinity of evaluation point E are selected as the target pixels, and the pixel values of these nine target pixels are determined according to the distance from evaluation point E. It is also possible to obtain a weighted average in which the weight is taken into consideration and use this as the feature amount x for the evaluation point E.

  As a general theory, when the feature quantity calculation unit 123 calculates the feature quantity x for a specific evaluation point E on a specific hierarchical image P, the specific evaluation is performed from the pixels constituting the specific hierarchical image P. A total of j pixels are extracted as the pixel of interest in order of proximity to the point E, and for the pixel values of the extracted j pixels of interest, weighting in consideration of the weight according to the distance between the specific evaluation point E and each pixel of interest An arithmetic operation for obtaining an average can be performed, and the value of the obtained weighted average can be used as the feature amount x.

<2.4 Modification of feature quantity extraction process>
Here, some modified examples of the procedure of the feature quantity extraction process described above will be described.

(1) Modification in which the differential image Dk is a hierarchical image In と し て 2.2, the processing of steps S33 to S36 shown in the flowchart of FIG. In this process, filter processing and reduction processing are alternately performed with the original image Q1 (first preparation image) as a starting point, and n images created by the filter processing according to the procedure as shown in FIG. Hereinafter, this process is a process of creating the image pyramid PP by adopting the filtered image as it is as the hierarchical images P1 to Pn.

  On the other hand, the modification described here is based on the processing of steps S33 to S36 shown in the flow chart of FIG. 7 and further, when the filtering processing of step S33 is completed, the kth obtained by the filtering processing The differential image “Pk−Qk” for subtracting the k-th preparation image Qk from the filtered image Pk (the image referred to as the k-th hierarchical image Pk in 2.22.2) is performed to calculate the k-th differential image Dk The required processing is added. In other words, in the modification described here, the processing of steps S33 to S36 shown in the flow chart of FIG. 7 is executed as it is, but further, the kth filtered image Pk to the kth preparation image Qk The difference operation "Pk-Qk" which subtracts 余 分 will be performed extra.

  Here, the difference operation “Pk−Qk” defines, for the k-th filtered image Pk and the k-th preparation image Qk, the pixels arranged at the same position on the pixel array as corresponding pixels, and on the image Pk The difference is obtained by subtracting the pixel value of the corresponding pixel on the image Qk from the pixel value of each pixel, and obtaining a difference image Dk consisting of a new set of pixels whose pixel value is the obtained difference.

  FIG. 23 is a plan view showing a procedure for creating an image pyramid PD composed of n kinds of difference images D1 to Dn by such difference operation “Pk−Qk”. Here, the first hierarchical image D1 shown on the upper right of the upper row is a difference image obtained by the difference operation “P1-Q1”. Specifically, the first filtered image P1 shown on the upper right of FIG. In FIG. 20, calculation is performed by difference operation (subtraction of pixel values of pixels at corresponding positions) by subtracting the first preparation image Q1 shown on the upper left side from the first layer image P1).

  Similarly, the second hierarchical image D2 shown in the middle right of FIG. 23 is a difference image obtained by the difference operation “P2-Q2”, and more specifically, the second filtering shown in the middle right of FIG. The difference is calculated by subtracting the second preparation image Q2 shown in the middle left from the image P2 (referred to as the second hierarchical image P2 in FIG. 20). The third hierarchical image D3 shown on the lower right of FIG. 23 is a difference image obtained by the difference operation “P3-Q3”. Specifically, the third filtered image shown on the lower right of FIG. It is calculated by a difference operation of subtracting the third preparation image Q3 shown on the lower left side from P3 (which is called the third hierarchical image P3 in FIG. 20). Hereinafter, the same difference calculation is performed, and finally, the difference image obtained by the difference calculation “Pn−Qn” is the n-th hierarchical image Dn.

  In the embodiment described in 2.2 2.2, as shown in FIG. 20, an image pyramid PP is formed by a first hierarchical image (first filtered image) P1 to an nth hierarchical image (nth filtered image) Pn. In the modification described here, as shown in FIG. 23, the image pyramid is formed by the first hierarchical image (first differential image) D1 to the nth hierarchical image (nth differential image) Dn. PD will be configured.

  As described above, in order to implement the modification described here, a procedure of difference calculation may be added to the procedure of the embodiment described in § 2.2. Specifically, the image pyramid creating unit 122 sets the original image created by the original image creating unit 121 as the first preparation image Q1, and is obtained by filtering the k-th preparation image Qk (where k is a natural number). A difference image Dk between the filtered image Pk to be reproduced and the k-th preparation image Qk is obtained, the difference image Dk is taken as a k-th hierarchical image Dk, and an image obtained by reducing the k-th filtered image The first hierarchical image D1 to the n-th hierarchical image Dn are included by alternately executing the filtering process and the reduction process until the n-th hierarchical image Dn is obtained as the preparation image Q (k + 1) of k + 1). An image pyramid PD consisting of a plurality n of hierarchical images may be created.

  After all, in this modification, the k-th hierarchical image Dk constituting the k-th hierarchical layer of the image pyramid PD is the image after the filter processing (filtered image Pk) and the image before the filter processing (preparation image Qk) It means a difference image, and the pixel value of each pixel corresponds to the difference between the pixel values before and after the filter processing. That is, while the k-th hierarchical image Pk in the embodiment described in 2.2 2.2 shows the image after filter processing itself, the k-th hierarchical image Dk in the modification described here is a filter. It will show the difference caused by the treatment. As described above, the image pyramid PP created in the embodiment described in 2.2 2.2 and the image pyramid PD created in the modification described here are significantly different in the meaning of the hierarchical image as the component thereof. As it turns out, there is no difference in that they are images that show some characteristics of the evaluation point E. Therefore, it is possible to extract the feature amount also from each hierarchy image D1 to Dn created in the modification described here, and the feature amount calculation unit 123 in the present modification includes each hierarchy image D1 to Dn. A process of extracting the feature amounts x1 to xn is performed.

(2) Modification Example of Creating Plural Image Pyramids by Plural Algorithms In the embodiment described in Section 2.2, as shown in FIG. 20, filtering is performed on the original image (first preparation image Q1). The image pyramid PP is created by an algorithm that determines each filtered image P1 to Pn by alternately executing the reduction process and reducing the filtered image P1 to Pn as n hierarchical images P1 to Pn. There is. On the other hand, in the modification in which the differential image Dk described in § 2.4 (1) is a hierarchical image, as shown in FIG. 23, each differential operation between the filtered image Pk and the preparation image Qk is performed. An image pyramid PD is created by an algorithm that obtains difference images D1 to Dn and sets the difference images D1 to Dn as n hierarchical images D1 to Dn.

  Also, there are various types of image filters used for filter processing as shown in FIG. 16 and FIG. 17, and various types of reduction processing (pooling processing) are also shown in FIG. 18 and FIG. 19. is there. As described above, even if the departure point is the same original image (the first preparation image Q1), the contents of each hierarchical image constituting the image pyramid obtained differ depending on the algorithm adopted. Moreover, in carrying out the invention of the prior application, it is not necessary to use only one image pyramid, and a plurality of image pyramids are created by a plurality of algorithms, and feature quantities are respectively extracted from the individual image pyramids. Is also possible.

  That is, the image pyramid creating unit 122 is provided with a function of performing image pyramid creating processing based on a plurality of different algorithms for one original image (first preparation image Q1), and a plurality of image pyramids are created. It may be done. In this case, the feature quantity calculation unit 123 determines, based on the pixel values of pixels (pixels located around the evaluation point) corresponding to the position of the evaluation point, for each hierarchical image constituting each of the plurality of image pyramids. A process of calculating the feature amount may be performed.

  For example, when the image pyramid creating unit 122 performs the image pyramid creating process, if the algorithm of the embodiment described in 2.2 2.2 is adopted as the main algorithm, as shown in FIG. A main image pyramid PP composed of P1 to Pn (filtered image) can be created, and as a sub-algorithm, an algorithm of a modification using the differential image Dk described in § 2.4 (1) as a hierarchical image If employed, as shown in FIG. 23, it is possible to create a sub-image pyramid PD composed of n sub-layer images D1 to Dn (difference images). Therefore, if the image pyramid creating unit 122 performs the image pyramid creating process using the above two types of algorithms, two image pyramids of the main image pyramid PP and the sub image pyramid PD can be created. .

  Here, the main image pyramid PP created by the filtering process using a Gaussian filter as illustrated in FIG. 14 can be called a Gaussian pyramid. Also, the sub-image pyramid PD configured using the difference image can be called a Laplacian pyramid. Since Gaussian pyramid and Laplacian pyramid become image pyramids whose characteristics are greatly different from each other, they are adopted as main image pyramid PP and sub image pyramid PD, and feature quantities are extracted using two image pyramids. If so, it becomes possible to extract more diverse feature quantities.

  On the other hand, the feature amount calculator 123 sets pixel values of pixels in the vicinity of the evaluation point E for the main layer images P1 to Pn constituting the main image pyramid PP and the sublayer images D1 to Dn constituting the sub image pyramid PD. If processing for calculating the feature amount is performed based on the feature amounts, the feature amounts xp1 to xpn calculated from the main layer images P1 to Pn and the feature amounts xd1 to xdn calculated from the sub layer images D1 to Dn can be obtained. . That is, a total of 2n feature quantities are extracted for one evaluation point E. In this case, since the feature amount for one evaluation point E is given as a 2n-dimensional vector to the estimation operation unit 132, more accurate estimation operation can be performed.

  After all, in order to implement the modified example described above, the image pyramid creating unit 122 sets the original image created by the original image creating unit 121 as the first preparation image Q1, and the kth preparation image Qk (where k is Let an image obtained by the filtering process for the natural numbers be the k-th main layer image Pk, and an image obtained by the reduction process for the k-th main layer image Pk be the (k + 1) th prepared image Q (k + 1). By alternately executing the filtering process and the reduction process until the main layer image Pn is obtained, a main image pyramid composed of n plural layer images including the first main layer image P1 to the n-th main layer image Pn The first sub-hierarchy is obtained by determining the difference image Dk between the k-th main layer image Pk and the k-th preparation image Qk, and setting the difference image Dk as the k-th sub-layer image Dk. It is sufficient to create a sub image pyramid consisting of hierarchy images of a plurality n as including a sub-hierarchy image Dn image D1~ the n.

  In addition, the feature amount calculation unit 123 may calculate the feature amount based on the pixel values of the pixels in the vicinity of the evaluation point E for each hierarchy image constituting the main image pyramid PP and the sub image pyramid PD. . Then, it becomes possible to extract a 2n-dimensional vector as a feature amount for one evaluation point E, and it is possible to perform more accurate estimation operation.

(3) Modification to create multiple original images and create multiple image pyramids In §2.4 (2) described above, by applying multiple algorithms to the same original image We have described a variant of creating multiple image pyramids. Here, a modified example will be described in which multiple original images are created to create multiple image pyramids.

  In the figure pattern shape estimation apparatus 100 'according to the prior invention, as shown in FIG. 1, the original image creation unit 121 creates an original image based on the given original graphic pattern 10. As the original image created here, as described in 2.1 2.1, the area density map M1 (FIG. 10), the edge length density map M2 (FIG. 11), the dose density map M3 (FIG. 13), etc. Various forms of images can be employed.

  In other words, when creating the original image based on the original figure pattern 10, the original image creation unit 121 can adopt various creation algorithms, and the contents differ depending on which algorithm is adopted. Various original images can be created. For example, the area density map M1 shown in FIG. 10, the edge length density map M2 shown in FIG. 11, and the dose density map M3 shown in FIG. 13 are images created based on the same original figure pattern 10, The pixel values of the individual pixels are different from one another, resulting in different images. Alternatively, based on the same original graphic pattern 10, prepare a plurality of density maps (in other words, a plurality of maps with different resolutions) different in pixel size (size of one pixel) and map size (number of pixels arranged in vertical and horizontal directions) Alternatively, a plurality of image pyramids may be created with each of the plurality of density maps as an original image. Of course, in theory, it is ideal to use a density map (density map with high resolution) with a small pixel size and a large map size as the original image, but since the computer has a finite memory, it is practical Preferably, a plurality of density maps having different pixel sizes and map sizes are used as the original image.

  Therefore, the original image creating unit 121 has a function of creating original images based on a plurality of different algorithms to create a plurality of original images, and the image pyramid creating unit 122 generates the plurality of original images. If it has a function of processing to create separate and independent image pyramids based on the images, multiple image pyramids can be created. Then, if the feature amount calculation unit 123 has a function of calculating the feature amount based on the pixel values of the pixels in the vicinity of the evaluation point E for each hierarchical image constituting each of the plurality of image pyramids, It becomes possible to extract feature quantities composed of higher dimensional vectors, and it is possible to carry out more accurate estimation operations.

  For example, the original image generation unit 121 generates a first original image including the area density map M1 illustrated in FIG. 10, a second original image including the edge length density map M2 illustrated in FIG. 11, and a dose density illustrated in FIG. With the function of creating a third original image consisting of the map M3, the image pyramid creating unit 122 creates three sets of independent image pyramids based on the three original images. Can. Both image pyramids are image pyramids created based on the same original graphic pattern 10. The feature amount calculation unit 123 can perform processing of calculating the feature amount based on the pixel values of the pixels in the vicinity of the evaluation point E for each hierarchical image constituting each of the three image pyramids. If n feature quantities x1 to xn are extracted from one image pyramid, a total of 3n feature quantities can be extracted for the same evaluation point E. That is, since a 3n-dimensional vector can be given as a feature amount for one evaluation point E, it is possible to perform more accurate estimation operation.

  Of course, it is also possible to combine the above-mentioned modification of 2.4 2.4 (2) with the modification described here. As described in 2.4 2.4 (2), if the image pyramid creating unit 122 adopts two different types of algorithms when creating the image pyramid, two ways of the main image pyramid PP and the sub image pyramid PD Can create an image pyramid. Therefore, if two image pyramids of the main image pyramid PP and the sub image pyramid PD are created based on each of the above three original images, a total of six image pyramids are obtained based on the same original graphic pattern 10. Can be created, and a 6n-dimensional vector can be given as a feature for one evaluation point E.

<<< 3. 3. Details of bias estimation unit >>>
Here, the detailed processing operation of the bias estimation unit 130 will be described. As shown in FIG. 1, the bias estimation unit 130 has a feature quantity input unit 131 and an estimation operation unit 132, and has a function of executing the process bias estimation process of step S4 in the flowchart of FIG. . In the case of the embodiment shown in FIG. 6, a feature amount (n-dimensional vector) of x1 to xn is input to the feature amount input unit 131 for one evaluation point E, and the estimation operation is performed by the estimation operation unit 132. An estimate y of the process bias for E is obtained. In the embodiment shown in FIG. 6, a neural network is used as the estimation operation unit 132. Therefore, the detailed configuration and operation of this neural network will be described below.

<3.1 Estimation operation by neural network>
In recent years, neural networks have attracted attention as a technology that forms the basis of artificial intelligence, and are used in various fields including image processing. The neural network is a computer construct that simulates the structure of the brain of an organism, and is composed of neurons and edges connecting them.

  FIG. 24 is a block diagram showing an embodiment in which a neural network is used as the estimation operation unit 132 shown in FIG. As illustrated, in the neural network, an input layer, an intermediate layer (hidden layer), and an output layer are defined, and predetermined information processing is performed on the information provided to the input layer in the intermediate layer (hidden layer), The result is output to the output layer. In the case of the prior invention, as shown, the feature quantities x1 to xn for one evaluation point E are given to the input layer as an n-dimensional vector, and the output layer estimates the process bias for the evaluation point E The value y is output. Here, as described in 11, the estimated value y of the process bias is an estimated value indicating the amount of deviation in the normal direction of the outline for the evaluation point E located on the outline of the predetermined figure. .

  In the case of the embodiment shown in FIG. 24, the estimation operation unit 132 has a neural network in which the feature amounts x1 to xn input by the feature amount input unit 131 are input layers and the estimated value y of process bias is an output layer. The middle layer of this neural network is composed of a first hidden layer, a second hidden layer,..., An Nth hidden layer which is the Nth hidden layer. These hidden layers have a large number of neurons (nodes), and edges connecting these neurons are defined.

  The feature quantities x1 to xn given to the input layer are transmitted as signals to each neuron through the edge. Finally, a signal corresponding to the estimated value y of the process bias is output from the output layer. Signals in the neural network are transmitted from the neurons of one hidden layer to the neurons of the next hidden layer through operations via edges. The calculation through the edge is performed using learning information L (specifically, parameters W and b described later) obtained at the learning stage.

  FIG. 25 is a diagram showing a specific operation process performed by the neural network shown in FIG. The portions shown by thick lines in the figure indicate the first hidden layer, the second hidden layer,..., The Nth hidden layer, and individual circles in each hidden layer connect neurons (nodes) and circles. The lines indicate the edges. As described above, feature quantities x1 to xn for one evaluation point E are given as n-dimensional vectors to the input layer, and an estimated value y of the process bias for the evaluation point E is a scalar for the output layer. It is output as a value (a dimensional value indicating the amount of deviation in the normal direction of the contour line).

  In the illustrated example, the first hidden layer is an M (1) -dimensional layer, and is constituted by a total of M (1) neurons h (1,1) to h (1, M (1)), and The two hidden layers are M (2) -dimensional layers, and are composed of a total of M (2) neurons h (2,1) to h (2, M (2)), and the Nth hidden layer is M (N) A layer of dimensions, consisting of a total of M (N) neurons h (N, 1) to h (N, M (N)).

  Here, the calculated values of the signals transmitted to the neurons h (1,1) to h (1, M (1)) of the first hidden layer are calculated using the same code as the calculated values h (1,1). Assuming that 表 す h (1, M (1)) is given, the values of these calculated values h (1, 1) to h (1, M (1)) are given by the matrix equation shown in the upper part of FIG. Be As the function f (ξ) on the right side of this equation, the sigmoid function shown in FIG. 27 (a), the normalized linear function ReLU shown in FIG. 27 (b), the normalized linear function Leakey ReLU shown in FIG. 27 (c), etc. The activation function of can be used.

  In addition, ξ described as an argument of the function f (ξ) is a matrix [W] and a matrix [x1 to xn] (feature amount given as an n-dimensional vector to the input layer, as shown in the middle part of FIG. It becomes the value which added matrix [b] to the product with). Here, the contents of the matrix [W] and the matrix [b] are as shown in the lower part of FIG. 26, and each component of the matrix (weighting parameter W (u, v) and biasing parameter b (u, v)) Is the learning information L obtained by the learning step described later. That is, the values of the individual components (parameters W (u, v), b (u, v)) constituting matrix [W] and matrix [b] are given as learning information L, as shown in FIG. If the computing equation is used, the computed values h (1,1) to h (1, M (1)) of the first hidden layer are calculated based on the feature quantities x1 to xn given to the input layer. Can.

  On the other hand, FIG. 28 is a diagram showing arithmetic expressions for obtaining respective values of the second to Nth hidden layers in the diagram shown in FIG. Specifically, the same sign is used for the operation value of the signal transmitted to the neurons h (i + 1, 1) to h (i + 1, M (i + 1)) of the (i + 1) hidden layer (1 ≦ i ≦ N) If the operation values h (i + 1, 1) to h (i + 1, M (i + 1)) are represented respectively, the values of these operation values h (i + 1, 1) to h (i + 1, M (i + 1)) are It is given by the matrix equation shown in the upper part of FIG. As the function f (ξ) on the right side of this equation, as described above, each function shown in FIG. 27 can be used.

  Further, ξ described as an argument of the function f (ξ) is the matrix [W] and the matrices [h (i, 1) to h (i, M (i))] as shown in the middle part of FIG. It becomes a value obtained by adding the matrix [b] to the product of (the operation value of neurons h (i, 1) to h (i, M (i)) of the immediately preceding i-th hidden layer). Here, the contents of the matrix [W] and the matrix [b] are as shown in the lower part of FIG. 28, and the individual components of the matrix (parameters W (u, v), b (u, v)) , And learning information L obtained by the learning step described later.

  Here too, the values of the individual components (parameters W (u, v), b (u, v)) that constitute matrix [W] and matrix [b] are given as learning information L, and FIG. By using the arithmetic expression shown, the arithmetic value of the (i + 1) th hidden layer is calculated based on the arithmetic values [h (i, 1) to h (i, M (i))] obtained in the ith hidden layer. h (i + 1, 1) to h (i + 1, M (i + 1)) can be calculated. Therefore, each value of the second to Nth hidden layers in the diagram shown in FIG. 25 can be sequentially obtained based on the arithmetic expression shown in FIG.

  FIG. 29 is a diagram showing an arithmetic expression for obtaining the value y of the output layer in the diagram shown in FIG. Specifically, the output value y (estimated value of the process bias for the evaluation point E: scalar value) is given by the matrix equation shown in the upper part of FIG. That is, the output value y is the matrix [W] and the matrix [h (N, 1) to h (N, M (N))] (neurons h (N, 1) to h (N, M of the Nth hidden layer) The product of (N)) and the scalar value b (N + 1) is added to the product of (N)). Here, the contents of the matrix [W] are as shown in the lower part of FIG. 29, and individual components (parameters W (u, v)) of the matrix [W] and the scalar value b (N + 1) Learning information L obtained by the learning stage.

  Thus, each value of the first hidden layer in the diagram shown in FIG. 25 is set to the parameters W (1, v), b prepared in advance as the learning information L in the feature quantities x1 to xn given as the input layer. Each value of the second hidden layer can be obtained by acting (1, v), and each value of the second hidden layer is set to a parameter W (2, v), which is prepared in advance as learning information L, to each value of the first hidden layer. Each value of the Nth hidden layer can be obtained by acting b (2, v), and each value of the Nth hidden layer is prepared in advance as learning information L to each value of the (N-1) th hidden layer The value of the output layer y can be obtained in advance as learning information L for each value of the Nth hidden layer by obtaining parameters W (N, v) and b (N, v) that are present. Parameter Parameters W (N + 1, v) and b (N + 1) You can ask for more. Specific arithmetic expressions are as shown in FIG. 26 to FIG.

  As described in (2) and (3) of 2.4 2.4, when adopting a modified example of creating a plurality of image pyramids, the feature quantity given as the input layer is an n-dimensional vector ( The vector of V · n dimensions (V is the total number of image pyramids) rather than x1 to xn) is obtained, but the number of input layers in the diagram shown in FIG. There is no change in the basic configuration and operation.

  Although the calculation for obtaining the estimated value y of the process bias for one evaluation point E using the neural network shown in FIG. 25 has been described above, actually, on the outline of the figure included in the original figure pattern 10 Similar operations are performed on a large number of defined evaluation points, and an estimated value y of process bias is obtained for each of the individual evaluation points. In the figure pattern shape correction apparatus 100 shown in FIG. 1, the pattern correction unit 140 corrects the pattern shape based on the estimated value y of the process bias for each evaluation point thus obtained (step S5 in FIG. 4). Will do.

<3.2 Learning stage of neural network>
As described above, in the case of the embodiment shown in FIG. 24, the estimation operation unit 132 is constituted by a neural network, and calculates the signal value transmitted to each neuron using the learning information L set in advance. It will be done. Here, the substance of the learning information L is the parameters W (u, v) and b (b (b), which are described in the lower part of FIG. 26, the lower part of FIG. 28, and the lower part of FIG. It is a value of u, v etc. Therefore, in order to construct such a neural network, it is necessary to obtain learning information L by the learning step executed in advance.

  That is, the neural network included in the estimation operation unit 132 is a dimension value obtained by actual dimension measurement of the actual figure pattern 20 actually formed on the actual substrate S by the lithography process using a large number of test pattern figures; Process bias estimation processing is performed using, as learning information L, parameters W (u, v), b (u, v), etc. obtained in the learning step using feature quantities obtained from each test pattern figure. become. Although the processing of the learning stage of such a neural network is a well-known technology, an outline of the processing suitable for the learning stage of the neural network used in the invention of the prior application will be briefly described here.

  FIG. 30 is a flow chart showing the procedure of the learning stage for obtaining the learning information L used by the neural network shown in FIG. First, in step S81, test pattern graphic creation processing is performed. Here, the test pattern figure corresponds to, for example, the original figure pattern 10 as shown in FIG. 2A, and a simple figure such as a rectangle or an L-shaped figure is usually used. In practice, thousands of test pattern figures having different sizes and shapes are created.

  Subsequently, in step S82, an evaluation point E is set on each test pattern graphic. Specifically, a process of defining a large number of evaluation points E at predetermined intervals may be performed on the outline of each test pattern graphic. Then, in step S83, feature amounts are extracted for each evaluation point E. The feature quantity extraction process of step S 83 is the same as the process described in § 2 and is executed by a unit having a function equivalent to that of the feature quantity extraction unit 120. According to the procedure described in 22, feature quantities x1 to xn are extracted for each evaluation point E.

  The estimation operation unit learning process of step S84 is a process of determining learning information L (that is, parameters W (u, v), b (u, v), etc.) using the feature quantity extracted in step S83. is there. In order to perform this learning process, the actual dimensions obtained by the actual lithography process are required. Therefore, in step S85, the lithography process is actually executed based on the test pattern figure created in step S81, and the actual substrate S is created. Then, in step S86, actual dimension measurement of individual figures is performed on the actual figure pattern formed on the actual substrate S. The measurement result is used for the learning process of step S84.

  Thus, the process of the learning stage shown in FIG. 30 is a process executed on a computer consisting of steps S81 to S84 (a process executed by a computer program) and a process executed on an actual substrate consisting of steps S85 and S86. And consists of The estimation calculation unit learning processing in step S84 determines the learning information L to be used for the neural network based on the feature amounts x1 to xn obtained by the processing on the computer and the actual dimensions measured on the actual substrate. It means processing.

  FIG. 31 is a flowchart showing a detailed procedure of learning of the estimation operation unit in step S84 in the flowchart shown in FIG. First, in step S841, the design position and feature amount of each evaluation point are input. Here, the design position of the evaluation point is the position on the test pattern graphic of the evaluation point set in step S82, and the feature amount of the evaluation point is the feature amounts x1 to xn extracted in step S83. Subsequently, in step S842, the actual position of each evaluation point is input. Here, the actual position of the evaluation point is determined based on the actual dimensions of each figure on the actual substrate S measured in step S86. Then, in step S843, the actual bias of each evaluation point is calculated. The actual bias corresponds to the amount of deviation between the design position of the evaluation point input in step S841 and the actual position of the evaluation point input in step S842.

  For example, when a rectangle like the original figure pattern 10 shown in FIG. 3A is created as a test pattern figure in step S81, evaluation points E11, E12, E13, etc. on the outline of this rectangle in step S82. Is set, and in step S83, the feature quantities x1 to xn are extracted for each of the evaluation points E11, E12, and E13. The positions and feature quantities of the evaluation points E11, E12, and E13 are input in step S841.

  On the other hand, a real figure pattern 20 as shown in FIG. 3B, for example, is formed on the real substrate S by the lithography process of step S85, and this real figure pattern 20 is constructed by real dimension measurement of step S86. The actual dimensions are measured for each side of the rectangle. Based on this measurement, the actual positions E21, E22, and E23 of the evaluation points are determined as points obtained by moving the evaluation points E11, E12, and E13 in the normal direction of the contour line. The actual positions E21, E22, E23 of each evaluation point are input in step S842.

  Therefore, in step S843, as shown in FIG. 3B, the actual bias y11, y12, y13 is calculated as the difference between the design position of each evaluation point E11, E12, E13 and the actual position E21, E22, E23. Ru. In practice, for example, the difference “b−” between the width b of the rectangle forming the real graphic pattern 20 shown in FIG. 3B and the width a of the rectangle forming the original graphic pattern 10 shown in FIG. The actual bias may be determined by an operation of obtaining a value y obtained by dividing a ′ ′ by 2. Of course, in practice, test pattern figures on the scale of several thousand are created, and a large number of evaluation points are set on the outline of each figure, so the procedure of steps S841 to S843 is a huge number of evaluation points. Will be executed for each. Here, focusing on one evaluation point E, for the evaluation point E, a combination of feature amounts x1 to xn and an actual bias y is prepared as a learning material.

  Subsequently, in step S844, the parameters W and b are set to initial values. Here, the parameters W and b are parameters W (u, v) and b (u (u, v) described in the lower part of FIG. 26, the lower part of FIG. 28, and the lower part of FIG. , V), etc., which are values constituting the learning information L. As an initial value, a random value may be given by a random number. In other words, in the initial stage before learning, the parameters W and b that constitute the learning information L have become non-functional values.

  Each step of steps S845 to S848 to be executed is an actual learning process, and by performing this learning process, the parameters W and b for which unnecessary values are initially set are gradually updated. Finally, it is corrected to an accurate value to the extent that it sufficiently functions as learning information L. First, in step S845, an operation of estimating the process bias y from the feature quantities x1 to xn is performed.

  Specifically, a neural network as shown in FIG. 24 is prepared. Of course, at this stage, random numbers are given as initial values to the parameters W and b that constitute the learning information L, and this neural network is incomplete because it can not fulfill its normal function as the estimation operation unit 132. It is a thing. The feature quantities x1 to xn input in step S841 are given to the input layer of this incomplete neural network, calculation is performed using learning information L consisting of incomplete values, and the estimated value y of the process bias is obtained as an output layer. calculate. Of course, the initially obtained estimated value y of the process bias is far from the observed actual bias.

  Therefore, in step S846, the residual to the actual bias at that time is calculated. That is, the difference between the estimated value y of the process bias obtained in step S845 and the calculated value of the actual bias obtained in step S843 is determined, and this difference is used as the residual for the evaluation point E. When this residual becomes less than a predetermined tolerance value, the learning information L at that time (ie, the parameters W (u, v), b (u, v), etc.) is learning information with sufficient practicality. It can be concluded that the learning phase can be ended.

  Step S847 is a procedure to determine whether or not the learning phase can be ended. In practice, since residuals are obtained for each of a large number of evaluation points, in practice, for example, if the amount of improvement of the residual sum of squares is less than a specified value, it is judged that learning should be ended. A determination method may be adopted. If it is not determined that the learning is ended, the parameters W (u, v), b (u, v) and the like are updated in step S848. Specifically, updating is performed to increase or decrease the value of each parameter W (u, v), b (u, v) or the like by a predetermined amount so as to produce an effect of reducing the residual. As a specific update method, methods based on various algorithms are known as learning methods in neural networks, so the explanation is omitted here, but as a basic policy, evaluation points of a large number of test pattern figures are An algorithm is employed that comprehensively takes into account the residuals and makes an overall adjustment so that each residual is totally reduced.

  Thus, the procedures of steps S845 to S848 are repeatedly executed until an affirmative determination is made in step S847. As a result, the values of the parameters W (u, v), b (u, v), etc. constituting the learning information L are gradually corrected in the direction of decreasing the residual, and finally, in step S847. An affirmative determination is made and the learning phase ends. The learning information L (parameters W (u, v), b (u, v), etc.) obtained at the end of learning is information suitable for obtaining an estimated value y of the process bias close to the actual bias in the output layer. It has become. Therefore, the neural network including the learning information L obtained at the learning end stage functions as the estimation operation unit 132 in the prior application invention.

<<< 4. 4. Method of estimating shape of graphic pattern according to prior application invention >>>
Up to this point, the invention of the prior application is regarded as a figure pattern shape estimation apparatus 100 'having a configuration shown in FIG. 1 or a figure pattern shape correction apparatus 100, and the configuration and operation thereof have been described. Here, the description of the prior application invention as a method invention of the figure pattern shape estimation method will be briefly described.

  When the invention of the prior application is understood as the invention of a method of estimating the shape of a figure pattern, the method estimates the shape of a real figure pattern formed on a real substrate by simulating a lithography process using the original figure pattern. It will be the way to do it. Then, in this method, the computer inputs an original figure pattern input step (the pattern created in step S1 in FIG. 4) in which the original figure pattern 10 including information of outlines indicating boundaries between the inside and outside of the figure is input. Step), an evaluation point setting step (step S2 in FIG. 4) in which the computer sets an evaluation point E at a predetermined position on the contour line of the input figure, and each evaluation of the original figure pattern 10 input by the computer. Based on the feature quantity extraction step (step S3 in FIG. 4) of extracting feature quantities x1 to xn indicating features around point E, and the original figure of each evaluation point E based on the feature quantities x1 to xn extracted by the computer A process bias estimation step (step S4 in FIG. 4) of estimating a process bias y indicating an amount of deviation between the position on the pattern 10 and the position on the real figure pattern 20; Constructed.

  Here, the feature quantity extraction stage (step S3 in FIG. 4) is an original image creation stage (figure in FIG. 4) for creating an original image Q1 consisting of a set of pixels U each having a predetermined pixel value based on the original figure pattern 10. 7 performs image pyramid creation processing including reduction processing (step S35 in FIG. 7) for creating a reduced image Qk (preparation image) based on the original image Q1 and a plurality of images having different sizes. Evaluation points E for the image pyramid creating step (steps S33 to S35 in FIG. 7) of creating the image pyramid PP composed of the hierarchy images P1 to Pn and each hierarchy image P1 to Pn constituting the created image pyramid PP And a feature amount calculating step (step S37 in FIG. 7) of calculating the feature amounts x1 to xn based on the pixel values of the pixels in the vicinity of.

  Here, in the process bias estimation stage (step S4 in FIG. 4), estimated values according to the feature quantities x1 to xn for the evaluation point E are calculated based on the learning information L obtained by the learning stage performed in advance. It includes an estimation operation step of determining and outputting the obtained estimated value as the estimated value y of the process bias for the evaluation point E.

  When the feature quantity extraction unit 120 described in 2 2 is used, a filter processing step of performing filter processing using a predetermined image processing filter on the original image Q1 or the reduced image Qk in the image pyramid creation step. By alternately executing (step S33 in FIG. 7) and a reduction process step (step S35 in FIG. 7) in which the reduction process is performed on the image Pk after the filter process, a plurality of hierarchical images P1 to Pn An image pyramid PP can be created.

  Specifically, in the procedure described in 2.2 2.2, at the image pyramid creation stage, an image in which the filtered image (filtered image Pk obtained in step S33 of FIG. 7) is used as hierarchical images P1 to Pn. A pyramid PP has been created. On the other hand, in the procedure of the modification described in 2.4 2.4 (1), at the image pyramid creation stage, the image after filter processing (filtered image Pk) and the image before filter processing (preparation image Qk) The differential image Dk is generated, and the image pyramid PD in which the generated differential images are hierarchical images D1 to Dn is generated.

<<< 5. 5. Basic configuration of shape correction device according to the present invention >>>
Now, in the sections 補正 1 to 4 4, the description has been given of the figure pattern shape correction apparatus according to the invention of the prior application. As described above, the technique according to the present invention has been developed to be applied to the figure pattern shape correction apparatus according to the prior invention. However, the present invention is not limited to the application to the prior invention. The present invention is characterized in that evaluation points are set by a unique method when simulating a lithography process, and shape correction is performed on the original figure pattern based on the result of simulation based on the evaluation points. Therefore, the present invention can be widely applied to general graphic pattern shape correction devices known from the prior art. However, for the sake of convenience of the description, a specific embodiment in which the present invention is applied to the figure pattern correction device of the prior application described in 1〜1 to 44 will be described for the convenience of explanation.

  The basic configuration of the figure pattern shape correction apparatus according to the embodiment described here is the same as the apparatus according to the prior invention described in 1〜1 to 44. That is, as shown in FIG. 1, the shape correction apparatus 100 described herein includes an evaluation point setting unit 110, a feature extraction unit 120, a bias estimation unit 130, and a pattern correction unit 140, and lithography based on the original figure pattern 10 is performed. When forming the actual figure pattern 20 on the actual substrate S by the process, the shape of the original figure pattern 10 is corrected so that the actual figure pattern 20 matches the original figure pattern 10, and the correction actually used in the lithography process It has a function of creating a figure pattern 15.

  In practice, the function as the shape correction device 100 of the graphic pattern is realized by incorporating a dedicated program into a general-purpose computer. Therefore, the evaluation point setting unit 110, the feature value extraction unit 120, the bias estimation unit 130, and the pattern correction unit 140 described above are each configured by incorporating a predetermined program in a computer.

  Here, the evaluation point setting unit 110, as described in ユ ニ ッ ト 1, relates to individual figures constituting the original figure pattern 10 (hereinafter referred to as a unit figure and denoted by the same reference numeral 10 as the original figure pattern). , And play a role in setting the evaluation point E at a predetermined position on the contour line indicating the boundary between the inside and the outside. Although FIG. 3A shows an example in which three evaluation points E11, E12, and E13 are set on a rectangular unit graphic 10, in fact, it goes around the outline of the unit graphic 10 one round. A number of evaluation points are set on the route to be The feature of the present invention lies in the method of setting this evaluation point. Specific evaluation point setting methods will be described in detail in §6.

  The feature quantity extraction unit 120 plays a role of extracting feature quantities x1 to xn indicating features around the individual evaluation points E in the original figure pattern 10. Section 2 shows the procedure using the image pyramids PP and PD as the feature value extraction method unique to the prior invention. Of course, even in the practice of the present invention, a method of extracting feature quantities using the image pyramids PP and PD can be adopted. However, the feature quantity extraction unit 120 used in the present invention is not limited to one that extracts feature quantities by the method described in §2. In the practice of the present invention, a unit for extracting feature amounts by an arbitrary method may be employed as the feature amount extraction unit 120. For example, instead of using the image pyramid PP consisting of plural n hierarchical images P1 to Pn, processing of calculating n characteristic quantities x1 to xn for each evaluation point E using plural n calculation functions is performed. It is also possible. As each calculation function, for the evaluation point E (X, Y) located at the coordinate value X, Y, by substituting the coordinate value X, Y as a variable, the evaluation point E (X, Y) and the original figure pattern A function (for example, a function used in the additional embodiment of the prior application) capable of calculating a predetermined function value by quantifying the positional relationship with each figure included in 10 may be used.

  The bias estimation unit 130 shifts the position between each evaluation point on the original figure pattern 10 and the position on the actual figure pattern 20 based on the feature amounts x1 to xn for each evaluation point extracted by the feature amount extraction unit 120. It plays a role in estimating the process bias y which indicates the quantity. Section 3 shows the procedure for estimating the process bias y by estimation operation using a neural network. Of course, the same procedure can be adopted to practice the present invention. However, the bias estimation unit 130 used in the present invention is not limited to one that estimates the process bias y by the method described in §3. In practicing the present invention, as the bias estimation unit 130, a unit that estimates the process bias y by any method may be adopted.

  The pattern correction unit 140 plays the role of creating the corrected figure pattern 15 by performing correction on the original figure pattern 10 based on the estimated value y of the process bias output from the bias estimation unit 130. The correction process by the pattern correction unit 140 is executed in consideration of the position of each evaluation point E set by the evaluation point setting unit 110 and the estimated value y of the process bias estimated for the evaluation point E. become. Therefore, specific correction processing according to the present invention will be described in detail in Section 7.

  After all, the shape correction apparatus 100 according to the present invention is the evaluation point setting unit 110, the feature value extraction unit 120, the bias estimation unit 130, and the pattern correction unit 140, similarly to the shape correction apparatus 100 according to the prior invention shown in FIG. The basic function of each of these components is the same as that of the device according to the invention of the prior application. However, the feature of the present invention resides in that the evaluation point setting unit 110 and the pattern correction unit 140 adopt unique processing functions. Hereinafter, this point will be described in detail.

<<< 6. 6. Details of evaluation point setting unit >>>
Here, the detailed configuration and the operation of the evaluation point setting unit 110 in the shape correction apparatus 100 according to the present invention will be described. FIG. 32 is a block diagram showing the configuration of the evaluation point setting unit 110 in the figure pattern shape correction apparatus 100 (see FIG. 1) according to the basic embodiment of the present invention. The basic function of the evaluation point setting unit 110 is to set an evaluation point E at a predetermined position on an outline indicating the boundary between the inside and the outside for each unit graphic forming the original graphic pattern 10.

  In order to realize such a basic function, the evaluation point setting unit 110 shown in FIG. 32 includes a unit graphic input unit 111, a unit graphic dividing unit 112, a divided graphic vector definition unit 113, a unit graphic vector definition unit 114, and a normal end point. A definition unit 115, an end point reconstruction unit 116, an evaluation point definition unit 117, and a replenishment end point definition unit 118 are provided. Each of these elements is actually configured by incorporating a predetermined program into a computer. Note that since the end point reconstruction unit 116 is not an essential component for the present invention, the end point reconstruction unit 116 can be omitted in practicing the present invention. However, in practice, it is possible to set the evaluation point more efficiently if the end point reconstruction unit 116 is provided. Therefore, hereinafter, a preferred embodiment in which the end point reconstruction unit 116 is provided will be described.

  As shown, the original figure pattern 10 and the maximum beam forming size Bmax are given as input to the evaluation point setting unit 110 shown here. Here, as shown in FIG. 1, the original figure pattern 10 is a figure pattern that is the basis of the lithography process, and is usually constituted by an aggregate of a large number of unit figures. On the other hand, the maximum beam forming size Bmax is the maximum beam forming size of the drawing apparatus used in the lithography process, and corresponds to the size of the maximum irradiation range by one shot of the light beam or electron beam irradiated by the drawing apparatus.

  On the other hand, from the evaluation point setting unit 110, the information of the contour line segment O sandwiched between the first end point T1 and the second end point T2 is output together with the information of the evaluation point E. Each of the end points T1 and T2 is a point defined on the outline of each unit graphic constituting the original graphic pattern 10, and is a point for defining both ends of the outline segment O. Further, in the case of the embodiment shown here, the evaluation point E is a point defined at the middle point of the contour line segment O. Although only one set of contour line segment O is drawn in FIG. 32, in practice, many contour line segments are set on the contour lines of each unit graphic, and evaluation points are set for each of the contour line segments. It will be done. Hereinafter, specific procedures for obtaining such an output will be described.

<6.1 Division of unit figure>
The unit graphic input unit 111 shown in FIG. 32 is a component for inputting each unit graphic forming the original graphic pattern 10. In the case of the embodiment described here, each unit graphic is constituted by a polygon, and FIG. 2A shows an example in which a rectangle is inputted as the original graphic pattern 10. The actual original graphic pattern 10 used to manufacture a semiconductor device or the like will be constituted by a complex graphic aggregate in which a large number of fine unit graphics having various shapes are collected. The unit figure is composed of polygons.

  The unit graphic division unit 112 shown in FIG. 32 executes a process of dividing each unit graphic input by the unit graphic input unit 111 into linear division lines to form a plurality of divided graphics. As described above, since each unit figure is a polygon, each divided figure obtained by dividing it by a straight dividing line also becomes a polygon. FIG. 33 is a plan view showing an example of the specific division processing performed by the unit graphic division unit 112. As shown in FIG. Here, for convenience of explanation, the case where the original figure pattern 10 consisting of one unit figure is input is taken as an example, and the unit figure is indicated by the same reference numeral “10” as the original figure pattern.

  FIG. 33 (a) is a plan view showing an example in which division by the unit graphic division unit 112 is performed on an elongated strip unit graphic 10 having a width of dimension Wsplit. Specifically, unit graphic input unit 111 inputs an elongated strip-shaped unit graphic 10 extending along a predetermined longitudinal axis (X axis in the illustrated example) and having linear upper and lower sides parallel to each other. On the assumption that the unit graphic division unit 112 divides the strip unit graphic 10 by a plurality of division lines (shown by broken lines in the figure) arranged at predetermined intervals along the longitudinal axis (X axis). A plurality of divided figures 10a, 10b, 10c, ..., 10j, 10k, 10l arranged one-dimensionally along the longitudinal axis (X axis) are formed (as described later, each divided figure The order of the signs a, b, c,... Is arranged from the left and right ends alternately toward the center).

  In the case of the illustrated example, dividing lines indicated by broken lines are arranged at intervals of the same dimension Wsplit as the width of the strip-shaped unit graphic 10 from each of the left end position and the right end position. Specifically, first, a split line is arranged at the position of Wsplit from the left end position indicated by the alternate long and short dash line to the right to define the divided graphic 10a, and then, the position of Wsplit from the right end position indicated by the alternate long and short dash line to the left And so on. Then, the dividing line is arranged at the position of Wsplit from the right dividing line of the divided figure 10a to the right, and the divided figure 10c is defined, and so on. The division lines are arranged alternately from the left and right ends toward the center. Then, when the width Wrest of the remaining portion left in the central portion becomes Wrest ≦ 2 × Wsplit, a dividing line for dividing the remaining portion into two is arranged to define divided graphics 10k and 10l. For this reason, the divided figures 10a to 10j become squares of which one side is Wsplit, and the central divided figures 10k and 10l become rectangles having a width of a fraction.

  In a semiconductor device or the like, an elongated strip-shaped unit graphic 10 as shown in FIG. 33A is often used as a pattern for forming a wiring layer or the like. Of course, the arrangement interval of the dividing lines does not necessarily have to match the vertical width Wsplit of the strip-shaped unit graphic 10, and may be set to any dimension. In this case, each divided figure is rectangular. As in the example shown in FIG. 33 (a), division is performed alternately from the left and right ends at a predetermined interval Wsplit, and finally, divided figures 10k and 10l having a width of Wrest / 2 are arranged at the center The reason for this is to avoid that a dividing line is placed in the immediate vicinity of the left and right ends of the unit graphic 10. As described later, since the evaluation point is set at the midpoint position of the side of each divided figure, when the horizontal width of the divided figures 10a and 10b at both ends shown in the figure becomes smaller, the evaluation point is closest to the vertex of the strip unit figure 10. It is not preferable because it will be set to If split by the method shown in FIG. 33 (a), the horizontal widths of the divided figures 10a and 10b at both ends can be made Wsplit, and the position separated by a distance Wsplit / 2 from the vertex (not close to the vertex but to some extent Evaluation points can be set at

  FIG. 33 (b) is a plan view showing an example in which division by the unit graphic division unit 112 is performed on a rectangular unit graphic 10 having a wider vertical width. As described above, when the unit graphic input unit 111 inputs the unit graphic 10 that spreads on the XY plane, the unit graphic dividing unit 112 first opens horizontal division lines parallel to the X axis at predetermined intervals in the Y axis direction. A plurality of elongated strip-shaped figures extending in the X-axis direction are formed by arranging a plurality of pieces and dividing the unit figure 10 by the plurality of horizontal division lines.

  In the example shown in FIG. 33 (b), horizontal dividing lines indicated by broken lines are arranged at intervals based on a predetermined dimension Wsplit from each of the upper end position and the lower end position. Specifically, first, a horizontal dividing line is disposed at the position of Wsplit below the upper end position indicated by the alternate long and short dash line to define the divided graphic 10a, and then at a position of Wsplit above the lower end position indicated by the alternate long and short dash line. A horizontal dividing line is arranged to define a divided figure 10b, and then a horizontal dividing line is arranged downward from the horizontal dividing line below the divided figure 10a at the position of Wsplit to define a divided figure 10c,. As such, the horizontal dividing lines are arranged alternately from the upper and lower ends toward the center. Then, when the vertical width Wrest of the remaining portion left in the central portion becomes Wrest ≦ 2 × Wsplit, a horizontal dividing line for dividing the remaining portion into two is arranged to define the divided graphics 10d and 10e. For this reason, elongated strip-shaped figures 10a, 10b, 10c having a longitudinal width Wsplit and elongated strip-shaped figures 10d, 10e having a vertical dimension having a fractional dimension are formed. As described above, it is also evaluated that divisions are alternately performed from the upper and lower ends at a width of a predetermined interval Wsplit and finally, the divided figures 10d and 10e having a width of Wrest / 2 are arranged in the center. This is to avoid that the point is set close to the vertex of the unit graphic 10.

  Thus, when the original rectangular unit graphic 10 is divided into a plurality of strip-shaped figures 10a, 10b, 10c, 10d and 10e, each of the plurality of strip-shaped figures 10a, 10b, 10c, 10d and 10e is further A plurality of divided figures (not shown) two-dimensionally arranged can be formed by dividing the plurality of divided vertical lines which are arranged at predetermined intervals along the X axis and parallel to the Y axis. That is, since each of the strip-shaped figures 10a, 10b, 10c, 10d, and 10e is an elongated figure extending in the X-axis direction as shown in FIG. 33 (a), as in the example shown in FIG. By dividing each in the horizontal direction, it is possible to obtain a plurality of divided figures two-dimensionally arranged. For example, if dividing by vertical dividing lines arranged at intervals of Wsplit along the X axis, it is possible to obtain a divided figure group including many squares of which one side is Wsplit.

  FIG. 33 (c) is a plan view showing an example in which the unit graphic division unit 112 divides the unit graphic 10 having an L-shaped band having a width Wsplit. Also in this example, if the same method as in FIG. 33 (b) is applied, first, horizontal split lines parallel to the X axis are arranged at intervals of Wsplit along the Y axis, and splitting by these horizontal split lines is performed . FIG. 33 (c) shows a state in which such division has been performed, and four sets of squares 10a, 10b, 10c, 10d and one set of elongated strip-shaped figures 10e are formed. Therefore, if the elongated strip-shaped figure 10e is further divided by a plurality of vertical dividing lines disposed along the X-axis at predetermined intervals (for example, Wsplit) and parallel to the Y-axis, the strip is formed similarly to FIG. The figure 10e can be divided into a plurality of divided figures (not shown). For example, if the strip 10e is divided by vertical dividing lines arranged at intervals of Wsplit along the X axis, it is possible to obtain a divided figure group in which squares of Wsplit are arranged in the X axis direction.

  As mentioned above, although the example of division about the unit figure which has the shape which consists of a rectangle or a combination of a rectangle was shown, the original figure pattern may include the unit figure which has an oblique side. FIG. 34 is a plan view showing an example of specific division processing performed on the parallelogram type of unit graphic 10 by the unit graphic division section 112 shown in FIG.

  Now, it is assumed that a unit figure 10 of parallelogram type as shown in FIG. 34 (a) is inputted by the unit figure input unit 111. The unit figure 10 has a parallelogram whose height is Wsplit and has a laterally elongated shape. The division method shown in FIG. 33 (a) described above can be applied to such parallelogram type unit graphic 10 as well. That is, as shown in FIG. 34 (b), unit graphic division unit 112 arranges division lines indicated by broken lines at intervals of dimension Wsplit, for example, from the left end position indicated by dashed dotted lines. Can be divided into four. As a result, the unit figure 10 is divided into four divided figures 10a, 10b, 10c and 10d, as shown in FIG. 34 (c).

  Here, the divided figures 10b and 10c positioned in the middle are squares of which one side is Wsplit, while the divided figures 10a and 10d positioned at both ends are trapezoids of which the height is Wsplit. By applying the division method shown in FIG. 33 (b) described above, it is possible to form a plurality of elongated strip-shaped figures extending in the X-axis direction by dividing the unit figure 10 of an arbitrary shape by a plurality of horizontal division lines. Finally, by dividing it further by the vertical dividing line, it is possible to form a divided figure consisting of a plurality of trapezoids. In practice, the unit graphic input unit 111 inputs a unit graphic 10 consisting of polygons, and the unit graphic division unit 112 divides this unit graphic 10 to form a divided graphic consisting of a trapezoid. Is preferred.

  Here, "trapezoidal" is generally defined as "a quadrilateral (one pair of opposite sides is parallel)" FIG. 35 (a) is a typical "trapezoid" meeting such a definition, and in this example, the height and the lower base are set to the dimension Wsplit. In FIG. 35 (b), one side is a Wsplit square, the upper base, the lower base, and the height are all set to the dimension Wsplit, and two pairs of opposite sides are parallel quadrilaterals. If the upper and lower bases are not of the size Wsplit, they are rectangular. On the other hand, FIG. 35 (c) shows a special trapezoid whose base is smaller than the dimension Wsplit and whose height is Wsplit, but whose upper base length is zero. Therefore, the term "trapezoid" in the present application broadly includes not only figures that fall under the definition of a general trapezoid, but also figures that fall under an extended definition that includes squares, rectangles, and triangles.

  As described above, by applying the dividing method shown in FIG. 33 (b), it is possible to divide a unit figure 10 consisting of an arbitrary polygon into divided figures consisting of "trapezoidal" as shown in FIG. Moreover, if the interval between the dividing lines is set to the predetermined interval Wsplit, as shown in FIGS. 35 (a), (b) and (c), a trapezoidal shape such that each has a vertical width Wsplit or less and a horizontal width Wsplit or less. It is possible to make divided figures. In carrying out the present invention, it is preferable to set the predetermined dimension Wsplit in the unit figure dividing unit 112 to form a trapezoidal divided figure having a vertical width Wsplit or less and a horizontal width Wsplit or less. The reason is explained below.

  The corrected figure pattern created by the figure correction apparatus 100 shown in FIG. 1 is used for an actual lithography process. Here, the actual lithography process includes an exposure step of irradiating a light beam or an electron beam onto the actual substrate S using an optical drawing apparatus or an electron beam drawing apparatus. In this exposure process, the maximum irradiable area that can be exposed by beam irradiation of one shot on the actual substrate S is determined by the maximum beam forming size Bmax of each drawing apparatus.

  For example, in a drawing apparatus capable of irradiating a trapezoidal shaped beam, when the maximum beam forming size Bmax is 20 nm, square irradiation of up to 20 nm on one side is performed in one-shot beam irradiation by the drawing apparatus. Only regions can be formed. Therefore, for example, when it is desired to perform exposure on a square region having a side of 30 nm, it is necessary to perform irradiation of a total of four shots while shifting the irradiation position in the longitudinal direction and the lateral direction. Taking these points into consideration, in order to perform the exposure process as efficiently as possible, each unit figure constituting the corrected figure pattern 15 used in the actual lithography process is also the value of the beam forming maximum size Bmax of the drawing apparatus. It is preferable to make the shape in consideration of

  Here, let us consider the case where shape correction is performed for a parallelogram unit figure 10 as shown in FIG. 34 (a). According to the dividing method described above, the unit graphic 10 is divided into four sets of divided figures 10a, 10b, 10c, and 10d as shown in FIG. 34 (c). At this time, if the unit figure dividing unit 112 is provided with a function to set the maximum beam forming size Bmax of the drawing apparatus used in the lithography process, the arrangement interval Wsplit of the dividing lines used for dividing the unit figure is the beam maximum forming It is possible to perform division processing set to the size Bmax or less.

  If the above-described dividing method is adopted, as shown in FIG. 35, each of the divided figures obtained by the dividing process becomes a trapezoid having a width Wsplit or less and a width Wsplit or less. Therefore, if the unit graphic dividing unit 112 sets the arrangement interval Wsplit of the dividing lines to the beam maximum molding size Bmax or less, the individual divided figures obtained by the division processing are all smaller than the vertical width Bmax and smaller than the horizontal width Bmax. It becomes trapezoidal. This is advantageous in improving the efficiency of the exposure process (reducing the number of total shots) when the actual lithography process is performed using the finally obtained corrected figure pattern 15.

  As will be described later, some of the sides of each of the divided figures obtained by the division process by the unit figure dividing unit 112 are adopted as contour line segments to be moved when the pattern correction unit 140 generates a corrected figure pattern, The contour line segment after the movement forms the contour of the corrected figure pattern.

  FIG. 36 is a plan view showing an example of a unit figure 15 after correction obtained by performing shape correction on each of the divided figures 10a, 10b, 10c and 10d constituting the unit figure 10 shown in FIG. 34 (c). is there. As shown, the divided figures 10a, 10b, 10c and 10d are respectively corrected to divided figures 15a, 15b, 15c and 15d, and the corrected figure 15a, 15b, 15c and 15d after correction A unit graphic 15 is configured. Specifically, in the case of this example, divided figures 15b and 15d whose vertical width is slightly reduced can be obtained by moving the upper side and lower side of divided figures 10b and 10d shown in FIG. ing.

  In the case of this example, the shape correction apparatus 100 shown in FIG. 1 corrects the unit graphic 10 shown in FIG. 34 (c) contained in the original graphic pattern 10, thereby obtaining the unit graphic after correction shown in FIG. By performing shape correction processing to create a corrected figure pattern 15 including 15, the corrected figure pattern 15 is used in an actual lithography process. Here, in this actual lithography process, consider what kind of exposure process is performed based on the unit figure 15 after correction shown in FIG.

  FIG. 37 is a plan view showing an example of an exposure process based on the unit figure 15 after correction shown in FIG. Here, a bold square represents one side of the irradiation area of the maximum beam forming size Bmax. In other words, the exposure apparatus used for the actual lithography process has a function of irradiating a light beam or an electron beam to the area within the thick square in one-shot irradiation. Of course, the actual irradiation area is only the part corresponding to the inner area of the unit graphic 15.

  Therefore, in the case of the example shown in FIG. 37, the exposure process based on the unit graphic 15 is completed by a total of four shots. That is, in the first shot, irradiation exposure is performed on the inner region of the hatched divided figure 15a in the squares of thick lines in FIG. 37 (a), and in the second shot, the irradiation in FIG. 37 (b) is performed. The irradiation exposure is performed on the inner region of the hatched divided figure 15b in the bold-lined squares, and in the third shot, the hatched divided figure 15c in the bold-lined squares of FIG. 37 (c). In the fourth shot, irradiation exposure is performed on the inner region of the hatched divided figure 15d in the squares of thick lines in FIG. 37 (d).

  As described above, the exposure process based on the unit figure 15 after correction is completed in four shots in total by causing the unit figure dividing unit 112 to form a trapezoidal divided figure having a vertical width Bmax or less and a horizontal width Bmax or less. This is because the pattern correction unit 140 is made to generate the unit graphic 15 after correction by performing correction to move the sides of the divided graphic as individual contour line segments.

  Of course, even if such division processing or correction processing is adopted, the effect of reducing the number of shots in the exposure step is not necessarily obtained. For example, in the example of FIG. 36, the divided figures 15b and 15d are figures in which the vertical widths of the divided figures 10b and 10d are slightly reduced, but conversely, in the case where the vertical figures are slightly expanded. The exposure steps of FIGS. 37 (b) and 37 (d) can not be performed in one shot. However, if the division process and the correction process described above are adopted, it is possible to obtain the corrected figure pattern 15 capable of improving the efficiency of the exposure process when viewed as the whole figure pattern including a large number of unit figures.

  In the block diagram of the evaluation point setting unit 110 shown in FIG. 32, although the beam maximum molding size Bmax is given to the unit figure dividing unit 112, this is a drawing apparatus used by the unit figure dividing unit 112 in an actual lithography process. It shows that it has a function of setting the beam maximum molding size Bmax and setting the arrangement interval of the dividing lines used for dividing the unit figure to the beam maximum molding size Bmax or less. Of course, such a function is not an essential function for carrying out the present invention, but in practice it is preferable to improve the efficiency of the exposure process as described above if such a function is provided. .

<6.2 Definition of vector and ordinary endpoints>
Subsequently, processing functions of the divided figure vector definition unit 113, the unit figure vector definition unit 114, and the normal end point definition unit 115 in the evaluation point setting unit 110 shown in FIG. 32 will be described. As described in 6.1 6.1, the unit graphic 10 is divided into a plurality of divided figures by the division process by the unit graphic division unit 112. Then, some of the sides of this divided figure are adopted as contour line segments to be moved at the time of correction.

  For example, in the case of the example shown in FIG. 34C, the unit graphic 10 is divided into four sets of divided figures 10a, 10b, 10c and 10d. Since each of these divided figures is a trapezoid having four sides, the total number of sides of the divided figures is 16. However, the total number of sides constituting the outline of the original unit graphic 10 is ten. Therefore, it is necessary to extract 10 sets of sides constituting an outline from the total of 16 sets of sides. The roles of the divided figure vector definition unit 113 and the unit figure vector definition unit 114 are thus to extract a specific side that constitutes the outline of the unit figure 10, and the role of the normal end point definition unit 115 is thus It is usually to define end points at both ends of each extracted side.

  FIG. 38 shows a divided figure vector definition unit 113, a unit figure vector definition unit 114, and a normal end point definition unit 115 shown in FIG. 32 with respect to the parallelogram unit figure 10 after division into four shown in FIG. It is a top view which shows the example in which the process by was performed. The processing by each of the definition units 113, 114, and 115 will be specifically described below, taking the parallelogram type unit graphic 10 as an example.

  First, the divided figure vector definition unit 113 makes one turn around a predetermined forward direction along the outline of the polygon forming the divided figure for each divided figure formed by the division process by the unit figure dividing unit 112. Then, the process of defining divided figure vectors on each side of the polygon is performed. For example, in the case of the example shown in FIG. 34 (c), four sets of divided figures 10a, 10b, 10c and 10d are formed. Therefore, for each divided figure, for example, if the clockwise direction is defined as the forward direction, as shown in FIG. 38 (a), the clockwise forward arrow A1 in each divided figure 10a, 10b, 10c, 10d. ~ A4 can be defined, and the divided figure vectors can be defined on the sides of each trapezoid so as to make one turn around the trapezoidal outline forming each divided figure in the direction indicated by the forward arrows A1 to A4. it can.

  FIG. 38 (b) is a plan view showing division vectors defined in this manner. For example, as shown in FIG. 38 (a), the trapezoid constituting the divided graphic 10a has four vertices ξ1 to ξ4, and the first side ξ1 to ξ2 and the second side と 2 to ξ3 , A third side 3-4 and a fourth side 4-1. Therefore, as shown in FIG. 38B, the divided graphic vector definition unit 113 performs processing of defining four sets of divided graphic vectors Va1, Va2, Va3 and V4a with respect to the divided graphic 10a.

  In other words, the divided figure vector Va1 is a vector along the first side ξ1-ξ2 with the vertex ξ1 as the start point and the vertex ξ2 as the end point, and the divided figure vector Va2 is the vertex ξ2 as the start point and the vertex ξ3 The end point is a vector along the second side ξ2-ξ3. The divided figure vector Va3 is a vector along the third side ξ3-ξ4 with the vertex ξ3 as a start point and the vertex ξ4 as an end point. The vector Va4 is a vector along the fourth side ξ4-ξ1 with the vertex ξ4 as a start point and the vertex ξ1 as an end point.

  Moreover, these four sets of vectors Va1, Va2, Va3 and Va4 are defined along the forward arrow A1 in the direction of circling the outline of the trapezoid making up the divided figure 10a, and the end point of one vector is It is in the relationship of starting point of the next vector. Therefore, an assembly of four sets of vectors Va1, Va2, Va3 and Va4 forms a vector group which makes one round in the clockwise direction the outline of the trapezoid constituting the divided graphic 10a.

  Of course, the divided graphic vector definition unit 113 may define the counterclockwise direction as the forward direction and perform the same processing. In this case, an assembly of four sets of vectors Va1, Va2, Va3 and Va4 forms a vector group which makes a round of the outline of the trapezoid forming the divided graphic 10a in a counterclockwise direction. As described above, whether the forward direction is clockwise or counterclockwise can be arbitrarily set, but one of the settings needs to be commonly applied to all divided figures. For example, in the example shown in FIG. 38 (a), all forward arrows A1 to A4 are clockwise arrows, but all of them may be counterclockwise arrows. However, the setting can not be changed for each divided figure such that the forward arrow A1 is clockwise, the forward arrow A2 is counterclockwise, and the like (to remove overlapping vector pairs as described later).

  FIG. 38 shows an example in the case where the divided figure is a square, but in general, the divided figure vector definition unit 113 follows the outline at each vertex of the divided figure made up of a K square. The first vertex to the Kth vertex are numbered in the order of clockwise or counterclockwise, and the kth vertex is a starting point, and the (k + 1) th vertex (however, the first vertex in the case of k = K) By repeatedly performing the process of defining the kth divided figure vector having the end point of k) up to k = 1 to K, the process of defining a divided figure vector for the divided figure is performed.

  Thus, as shown in FIG. 38B, a total of 16 sets of divided graphic vectors Va1 to Vd4 are defined by the divided graphic vector definition unit 113 on each of the four sides of each of the divided graphic 10a, 10b, 10c, 10d. It will be In FIG. 38 (b), for convenience of illustration, two vectors defined on the same side are drawn slightly offset from the position of the side, but in actuality, these two vectors are drawn. Is a vector defined redundantly on the same side. For example, vectors Va2 and Vb4 shown in FIG. 38 (b) are vectors defined on side ξ2-ξ3.

  From the set of divided figure vectors thus defined, unit figure vector definition unit 114 searches for overlapping vector pairs which are arranged on overlapping sides of a pair of adjacent polygons adjacent to each other and whose directions are opposite to each other, A process of defining a plurality of unit graphic vectors arranged along the outline of the unit graphic 10 is performed by removing the overlapping interval part of the searched overlapping vector pairs and using the remaining vectors.

  For example, in the case of the example shown in FIG. 38 (b), since a total of 16 divided figure vectors Va1 to Vd4 are defined, an overlapping vector pair is searched from among these 16 divided figure vectors Va1 to Vd4. A process is performed to remove the overlapping interval part of the searched overlapping vector pair. Here, as described above, the overlapping vector pair is “a pair of vectors which are disposed on overlapping sides of a pair of adjacent polygons adjacent to each other and whose directions are opposite to each other”. In the example shown in b), the vectors Va2 and Vb4 are searched as the first overlapping vector pair, the vectors Vb2 and Vc4 are searched as the second overlapping vector pair, and the vectors Vc2 and Vd4 are detected as the third overlapping vector pair It is searched.

  Since these respective overlapping vector pairs overlap in all sections, the process of removing them completely is performed. FIG. 38 (c) is a plan view showing the state after such removal processing is performed, and the overlapping vector pairs (Va2, Vb4), (Vb2, Vc4), (Vc2) shown in FIG. 38 (b) are shown. , Vd4) are all removed. Here, the 16 sets of vectors Va1 to Vd4 shown in FIG. 38 (b) are referred to as “divided figure vectors”, and therefore 10 sets of vectors shown in FIG. 38 (c) are 16 shown in FIG. 38 (b). In order to distinguish it from a set of vectors, it will be called "unit figure vector", and the signs will be changed to V1 to V10. Therefore, unit figure vectors V1 to V10 shown in FIG. 38 (c) are actually the same as divided figure vectors Va1, Vb1,..., Va3 and Va4 shown in FIG. 38 (b).

  The unit figure vectors V1 to V10 thus obtained are vectors arranged on the outline of the unit figure 10, and are defined in a direction in which the outline is made to go around in the forward direction (clockwise direction in the example shown). The end point of one vector is in the relation of being the start point of the next vector. Therefore, a set of ten unit figure vectors V1 to V10 forms a vector group that goes around the outline of the unit figure 10 clockwise.

  The normal end point definition unit 115 performs processing for defining the normal end point at each of the start positions (the same as in the end point position) of the unit graphic vectors thus defined. FIG. 38 (d) is a plan view showing a state in which normal end points A to J (indicated by x in the figure) are defined at the start positions of the 10 sets of unit graphic vectors V1 to V10 shown in FIG. is there. Each of the normal end points A to J thus defined is, among the vertices of the divided figures 10a, 10b, 10c and 10d shown in FIG. . In other words, each of the normal end points A to J is an end point that delimits both ends of the side located on the outline of the unit graphic 10 among the sides of each divided graphic. Note that the end point defined in this manner is referred to as a “normal end point” for convenience of distinction from the “replenishment end point” additionally defined by the replenishment end point definition unit 118 described later.

  On the other hand, in FIG. 39, the process by the unit figure dividing unit 112, the divided figure vector defining unit 113, the unit figure vector defining unit 114, and the normal end point defining unit 115 shown in FIG. It is a top view which shows the said example. When an L-shaped unit graphic 10 as shown in FIG. 39 (a) is input by the unit graphic input unit 111, division processing as shown in FIG. 39 (b) is performed by the unit graphic dividing unit 112. Three sets of divided figures 10a, 10b and 10c are formed.

  Then, for these divided figures 10a, 10b and 10c, forward arrows A1 to A3 are set in the clockwise direction as shown, and the divided figure vector definition unit 113 adds up the total as shown in FIG. 39 (c). Twelve sets of divided graphic vectors Va1 to Vc4 are defined. Subsequently, the unit graphic vector definition unit 114 searches for an overlapping vector pair, and performs processing for removing an overlapping interval part of the searched overlapping vector pair. In the case of the example shown in FIG. 38 (b) described above, the entire interval of the searched overlapping vector pair is the overlapping interval part, so the overlapping vector pair is completely eliminated, but FIG. 39 (c) In the example shown in, there exist overlapping vector pairs in which only a part of them overlap.

  Specifically, in the case of the example shown in FIG. 39C, the vectors Va3 and Vb1 are searched as the first overlapping vector pair, and the vectors Vb2 and Vc4 are searched as the second overlapping vector pair. Here, since the second overlap vector pair Vb2 and Vc4 is an overlap interval part in the entire interval, processing is performed to remove them entirely. However, the first overlap vector pair Va3 and Vb1 are shown in FIG. As shown in the upper part of (d), only a part is an overlapping section part (the part shown by a thick line), so processing is performed to remove only the overlapping section part. As a result, as shown in the lower part of FIG. 39 (d), only the non-overlapping interval portion remains as the unit graphic vector V3.

  FIG. 39 (e) is a plan view showing the state after the above-described removal processing is performed by unit graphic vector definition unit 114, and shows the state in which nine sets of unit graphic vectors V1 to V9 remain. It is done. The unit figure vectors V1 to V9 obtained in this way are vectors disposed on the outline of the L-shaped unit figure 10 shown in FIG. 39 (a), and in the forward direction (in the example shown, clockwise direction ) Is defined in the direction in which the contour line goes around, and the end point of one vector is in the relation of being the start point of the next vector. Therefore, an assembly of these nine sets of unit graphic vectors V1 to V9 constitutes a vector group which makes a round of the outline of the unit graphic 10 in the clockwise direction.

  The normal end point definition unit 115 performs processing of defining a normal end point at each of the start positions of the unit graphic vectors thus defined. FIG. 39 (f) is a plan view showing a state where normal end points A to I (indicated by x in the figure) are defined at the start positions of the nine unit graphic vector V1 to V9 shown in FIG. 39 (e). is there.

  Subsequently, an example of processing for a so-called punched unit graphic having an area outside the graphic inside will be described. In FIG. 40, the unit graphic division unit 112, divided graphic vector definition unit 113, unit graphic vector definition unit 114, and normal end point definition unit 115 shown in FIG. It is a plan view showing an example. This example is an example in the case where a B-shaped unit graphic 10 as shown in FIG. 40A is input by the unit graphic input unit 111. In the case of the B-shaped unit graphic 10 shown in the figure, the figure main body 10X is a hatched area, but a rectangular hole opening 10Y exists inside. That is, the unit figure 10 of such a hole opening will have the outer outline Oout and the inner outline Oin. Even when such a perforated unit graphic 10 is input, basically, the same processing as the processing described above may be executed.

  First, division processing as shown in FIG. 40 (b) is performed by the unit graphic division unit 112 to form eight sets of divided graphics 10a to 10h. Then, for these divided figures 10a to 10h, as shown in FIG. 40 (c), forward arrows A1 to A8 are set in the clockwise direction, and the divided figure vector definition unit 113 shows them in FIG. 40 (d). Thus, a total of 32 sets of divided graphic vectors Va1 to Vh4 are defined. Subsequently, the unit graphic vector definition unit 114 searches for an overlapping vector pair, and performs processing for removing an overlapping interval part of the searched overlapping vector pair.

  FIG. 40 (e) is a plan view showing the state after the above-described removal processing has been performed by unit graphic vector definition unit 114, showing the state in which 20 sets of unit graphic vectors V1 to V20 remain. There is. The point to be noted here is that the unit figure vectors V1 to V12 are vectors disposed on the outer outline Oout of the unit graphic 10 in the shape of the letter B shown in FIG. In the case of (1), the unit graphic vectors V13 to V20 are defined in the direction of circling the outline in the clockwise direction), whereas the unit graphic vectors V13 to V20 are shown in FIG. It is a vector disposed on the inner contour Oin, and is a point defined in a reverse direction (in the illustrated example, in the counterclockwise direction) so as to go around the contour.

  In the present application, with respect to such a unit figure of a hole, the figure forming the outer outline Oout is referred to as a "positive figure", and the figure forming the inner outline Oin is referred to as an "anti-figure". In the illustrated example, unit figure vectors V1 to V12 are a group of vectors that make a round in the forward direction of a positive figure, and unit graphics vectors V13 to V20 are a group of vectors that make a round in the opposite direction. Therefore, in the case of a unit figure having holes, if the vector group makes a round in the forward direction along the outline, it can be determined that the outline is a regular figure that constitutes the outer outline Oout of the unit figure, and the vector group is In the case of going around in the opposite direction along the outline, it can be determined that the outline is the opposite figure that constitutes the inner outline Oin of the unit figure. Such differences between positive and negative figures can be used in the end point reconstruction process described in 6.3 6.3.

  The normal end point definition unit 115 performs processing of defining a normal end point at each of the start positions of the unit graphic vectors thus defined. FIG. 40 (f) is a plan view showing a state in which normal end points A to T (indicated by x in the figure) are defined at the start points of the 20 sets of unit graphic vectors V1 to V20 shown in FIG. 40 (e). is there. As shown in the figure, normal end points are respectively defined on both the outer contour Oout and the inner contour Oin of the perforated unit graphic 10.

<6.3 Reconstruction of Endpoints>
Subsequently, the function of the end point reconstruction unit 116 in the evaluation point setting unit 110 shown in FIG. 32 will be described. The end point reconstruction unit 116 is a component that performs reconstruction to correct the number or the position of the normal end points defined on the outline of the unit graphic by the end point definition unit 115. As described in 6.2 6.2, since the end point definition unit 115 usually defines the end point based on the vertex position of each divided figure, the division process by the unit figure division unit 112 is a line appropriately. If so, the number and location of the defined normal endpoints should also be appropriate. However, in practice, it is possible to set evaluation points more efficiently by adjusting the number and position of the normal end points thus defined.

  Therefore, in the case of the embodiment described here, the end point restructuring unit 116 reconstructs the normal end point defined by the normal end point definition unit 115. Of course, such a reconstruction process is not an essential process for the present invention. Therefore, the end point reconstruction unit 116 can be omitted in practicing the present invention. However, in practice, providing the end point reconstruction unit 116 enables more efficient evaluation point setting. The end point reconstruction unit 116 reconstructs the normal endpoints defined by the endpoint definition unit 115 by reducing their number or correcting their positions. Hereinafter, several methods for carrying out the reconstruction will be specifically described.

(1) Reconstruction by Three-Point End Point Removal Processing FIG. 41 is a schematic diagram showing a procedure in the case where three-point end point removal processing is adopted as the end point reconstruction processing by the end point reconstruction unit 116 shown in FIG. (FIG. 41 (a)) and the flowchart (FIG. 41 (b)). The end point removal processing is usually a part or all of the normal end points (in fact, normal end points that do not constitute the vertex of the polygon that constitutes the unit figure) for some or all of the unit figures defined by the end point definition unit 115 Can be said to be processing to perform reconstruction by reducing the number of end points.

  As described in 6.2 6.2, the normal end points defined by the normal end point definition unit 115 are defined in a predetermined order on the outline of the unit graphic 10. For example, in the case of the example shown in FIG. 38 (d), the normal end points are defined in the order of A to J on the outline of the unit figure 10, and in the case of the example shown in FIG. Normally, end points are defined in the order of A to I on the line, and in the case shown in FIG. 40 (f), A to L on the outer contour Oout of the unit graphic 10 and M to T on the inner contour Oin. The endpoints are usually defined in the following order. The three-point end point removal process described here determines whether or not to remove the normal end point to be determined with reference to the distance between the previous normal end point and the one subsequent normal end point. The method is adopted.

  Now, a total of N normal end points T (1) to T (N) are defined on one contour line of the unit graphic, and the nth normal end point T (n) among them is determined as the normal end point. Consider the case of determining whether or not T (n) should be removed. In this case, as shown in FIG. 41 (a), the distance between the n-th normal end point T (n) to be subjected to removal determination and the immediately preceding n-th normal point T (n-1). A distance L2 between L1 and the n-th normal end point T (n) to be removed and determined and the (n + 1) -th normal end point T (n + 1) after it is determined.

  However, there are some points to be noted about the distance L1. For example, in FIG. 41 (a), when the normal end point T (n-1) has already been removed, the distance between the immediately preceding normal end point T (n-2) and T (n) Let L1 be. Further, when the normal end point T (n-2) is also to be removed, the distance between the immediately preceding normal end point T (n-3) and T (n) is L1. After all, the distance L1 here is, in general terms, the n-th normal end point T (n) to be subject to removal determination and the normal end point T (n-m) not immediately prior to it and not to be removed (However, m is a natural number, and in the case of n−m <1, it is defined as the distance to T (n−m + N)). Thus, after the distances L1 and L2 are defined, the n-th normal end point T (n) is determined to be a removal target when both of the following two determination conditions are satisfied.

  The first determination condition is a condition that “at least one of L1 and L2 is less than or equal to a predetermined reference value Wmerge”. Here, Wmerge is a predetermined reference value set in advance, and in general, it is preferable to set a smaller value than Wsplit used by the unit graphic dividing unit 112 for the dividing process. The reference value Wmerge removes the normal end point T (n) when the normal end point T (n) to be subjected to the removal determination is close to the adjacent normal end point T (n−m) or T (n + 1) to some extent This is a value that defines the critical approach distance when determining that the end point T (n-m) or T (n + 1) should be fused.

  If the reference value Wmerge is set to a large value, the possibility that it is determined that the end point T (n) should normally be removed is high, and if it is set to a small value, the possibility is low. Of course, if the possibility of being determined to be removed is high, the computational burden is reduced but the accuracy of the simulation is lowered, and if the possibility is low, the accuracy of the simulation is improved but the computational load is large. Become. Therefore, in practice, the reference value Wmerge may be set to an appropriate value in consideration of the balance between the maintenance of the simulation accuracy and the reduction of the calculation load.

  The second determination condition is a condition that “normally three points of the end points T (n−1), T (n), T (n + 1) are on a straight line”. FIG. 41 (a) shows a state in which such a condition is satisfied. For example, when the end point T (n) is the vertex of a unit figure, the end point T (n-1) The condition is not satisfied because T (n + 1) is an end point on a different side. Therefore, this second determination condition is the condition that "the three points are on the same side". In other words, if the normal end point that does not constitute the vertex of the polygon that constitutes the unit graphic is called the intermediate normal end point, the second determination condition is that “the normal end point T (n) to be determined for removal is It is a condition that it is an intermediate normal end point.

  FIG. 41 (b) is a flow chart showing a procedure of end point reconstruction based on such three-point end point removal processing. First, in step SS1, a parameter n indicating the number of a normal end point to be subjected to removal determination is set to n = 1, and in step SS2, it is determined whether L1 ≦ Wmerge. If a negative determination is made here, it is determined in step SS3 whether L2 ≦ Wmerge. If a positive determination is made in any of steps SS2 and SS3, then in step SS3, whether three points T (n-1), T (n) and T (n + 1) lie on a straight line. Is determined, and a positive determination is made, processing for normal end point T (n) to be removed is performed in step SS5. On the other hand, if a negative determination is made in step SS3 or SS4, in step SS6, processing for excluding the normal end point T (n) as a removal target is performed.

  Such processing is repeatedly performed until n = N is reached through steps SS7 and SS8 while the parameter n is increased by one. Then, when n = N is reached, the process proceeds from step SS7 to SS9, and the normal end points to be removed are collectively removed.

  In the above processing procedure, when n = 1, the removal determination target is the first normal end point T (1), so the normal end point T (n-1) in each of the above determination conditions is the Nth N It is necessary to read it as the normal end point T (N) of. Similarly, in the case of n = N, the removal determination target is the Nth normal end point T (N), so the normal end point T (n + 1) in each of the above determination conditions is the first normal end point T (1). It needs to be read as).

  After all, the end point reconstructing unit 116 can calculate the total N defined for one contour line of one unit graphic, as a general theory, in the case of adopting the reconstruction by the end point removal processing of the three-point type described here. The ordinary end points are referred to as first ordinary end point T (1) to N-th ordinary end point T (N) in order along the contour line, and focusing on the n-th ordinary end point T (n) , The (n-1) th normal end point T (n-1) (where, if n = 1, the Nth normal end point T (N): the same applies hereinafter) and the nth normal end point T (n) And the n-th normal end point T (n) and the (n + 1) -th normal end point T (n + 1) (however, the first normal end point T (1) in the case of n = N: the same applies hereinafter) And at least one of L1 and L2 is equal to or less than a predetermined reference value Wmerge, and the normal end point T -1) Repeatedly executes the judgment processing for removing the n-th normal end point T (n) to n = 1 to N when the three points T (n) and T (n + 1) are on a straight line End point removal processing for removing the normal end point to be removed.

  FIG. 42 applies the three-point end point removal process shown in FIG. 41 to the normal end points A to J on the parallelogram type unit graphic 10 shown in FIG. 38 (d) to reconstruct the end points It is a top view which shows the example performed. Here, FIG. 42 (a) shows a state in which the end points D and I (indicated by white triangles in the figure) are usually determined to be removed before reconstruction, and FIG. 42 (b) shows the state. The state after reconstruction is shown in which the normal end points D and I determined to be removed are removed.

  In the example shown in FIG. 42, the reason why the normal end point D is determined to be removed is that the first determination condition of “Lde ≦ Wmerge” is satisfied for the distance Lde between the end points D and E, and “three end points This is because the second determination condition that C, D, and E are on a straight line is satisfied. Similarly, the reason why the normal end point I is determined to be removed is that the first judgment condition of “Lij ≦ Wmerge” is satisfied for the distance Lij between the end points I and J, and “three end points H, I, This is because the second determination condition that “J is on a straight line” is satisfied. In the end, the end point reconstruction process merges the end point D to the end point E, the end point I to the end point J, and the number of ten normal end points to eight.

  The removal of the end points D and I slightly lowers the accuracy of the simulation to be performed later, but since the distances Lde and Lij are both less than or equal to the reference value Wmerge, there is no serious problem in practical use. On the other hand, the removal of the end points D and I reduces the computational load of the simulation.

  FIG. 43 applies the three-point end point removal process shown in FIG. 41 to the normal end points A to I on the L-shaped unit graphic 10 shown in FIG. It is a plan view showing an example. Here, FIG. 43 (a) shows a state in which the end point D (indicated by a white triangle in the figure) is usually determined as a removal target before reconstruction, and FIG. 43 (b) shows this state. The state after the reconstruction which removed the normal end point D determined as removal object is shown.

  In the example shown in FIG. 43, the reason why the normal endpoint D is determined to be removed is that the first determination condition of “Lcd ≦ Wmerge” is satisfied for the distance Lcd between the endpoints C and D, and “three endpoints This is because the second determination condition that C, D, and E are on a straight line is satisfied. In the end, the end point reconstruction process merges the end point D to the end point C, and reduces the nine normal end points to eight. After all, although this reconstruction slightly lowers the accuracy of the simulation to be performed later, since the distance Lcd is equal to or less than the reference value Wmerge, there is no problem in practical use. On the other hand, removal of the end point D reduces the computational load of the simulation.

(2) Reconstruction by Four-Point Type Endpoint Removal Processing FIG. 44 is a schematic diagram showing a procedure in the case where four-point type endpoint removal processing is adopted as the endpoint reconstruction processing by the endpoint reconstruction unit 116 shown in FIG. (FIG. 44 (a)) and a flowchart (FIG. 44 (b)). The end point removal processing described here, like the end point removal processing of the three-point system described above, removes a part or all of the intermediate normal end points for a part or all of the unit graphics defined by the end point definition unit 115 normally. This process is usually a process that performs reconstruction by reducing the number of end points. However, the four-point end point removal processing described here is intended to be the removal determination target of two adjacent sets of normal end points, and for the two sets of normal end points targeted for this removal determination, the previous end points In this method, a method of determining which one should be removed together with the determination of whether or not to remove it is adopted with reference to the distance between the point and the next normal end point.

  Now, a total of N normal end points T (1) to T (N) are defined on one contour line of the unit figure, and the nth normal end point T (n) and the (n + 1) th normal end point among them Consider a case in which two points of T (n + 1) are to be judged, and it is judged whether or not the judgment object should be removed, and which of the two points should be removed if it is to be eliminated. In this case, as shown in FIG. 44 (a), the distance L2 between two points of the n-th ordinary end point T (n) and the (n + 1) -th ordinary end point T (n + 1) to be removed and determined If both of the two determination conditions are satisfied, it is determined that one of them is to be removed.

  The first determination condition is a condition that “L2 is less than or equal to a predetermined reference value Wmerge”. Here, Wmerge is a predetermined reference value set in advance, as in the case of the three-point end point removal process described above, and in general, is smaller than Wsplit which unit graphic division unit 112 uses for the division process. It is preferable to set a value. The reference value Wmerge determines that if two sets of normal end points T (n) and T (n + 1) to be subject to removal determination approach each other to a certain extent, one should be eliminated to merge the two. Is a value that determines the critical approach distance when Therefore, the reference value Wmerge may be set to an appropriate value in consideration of the balance between the accuracy maintenance of the simulation and the reduction of the calculation load, as in the end point removal process of the three-point system.

  The second determination condition is the (n-1) th normal endpoint T (n-1) immediately preceding the n-th normal endpoint T (n) and the (n + 1) th normal endpoint T (n + 1). In consideration of the next (n + 2) normal end point T (n + 2), “normal end points T (n−1), T (n), T (n + 1), T (n + 2) are one straight It is the condition of being on the line. FIG. 44 (a) shows a state in which such a condition is satisfied. For example, when the end point T (n) is the vertex of the unit figure, the normal end point T (n-1) The condition is not satisfied because T (n + 1) is an end point on a different side. Therefore, this second determination condition is the condition that "the four points are on the same side". In other words, if the normal end that does not constitute the vertex of the polygon that constitutes the unit graphic is called the intermediate normal end, the second determination condition is that “the normal end T (n) to be determined for removal and The condition is that T (n + 1) is an intermediate normal end point.

  If both of the above two determination conditions are satisfied, one of the end points T (n) and T (n + 1) is usually determined to be a removal target. Here, the distance between the end points T (n) and T (n-1) is usually L1, and the distance between the end points T (n + 1) and T (n + 2) is L3. Sometimes, it may be determined based on the magnitude relationship between the distances L1 and L3. That is, in the case of L1 <L3, the end point T (n) is the removal target, and in the case of L1> L3, the normal end point T (n + 1) is the removal target. Then, since the distance between two end points (L1 + L2 in the former case and L2 + L3 in the latter case) sandwiching the end points to be removed can be made as small as possible, the influence of the reduction in simulation accuracy due to the removal process can be made as much as possible. It can be made smaller. In the case of L1 = L3, either of the normal end points T (n) and T (n + 1) may be the removal target. In the case where the four points are not on a straight line but the three points are on a straight line, a method according to the end point removal process of the three-point system described above may be executed.

  Also here, there is a point to be slightly noted about the distance L1. For example, in FIG. 44 (a), if the normal end point T (n-1) has already been removed, the distance between the immediately preceding normal end point T (n-2) and T (n) Let L1 be. Further, when the normal end point T (n-2) is also to be removed, the distance between the immediately preceding normal end point T (n-3) and T (n) is L1. After all, the distance L1 here is, in general terms, the n-th normal end point T (n) to be subject to removal determination and the normal end point T (n-m) not immediately prior to it and not to be removed (However, m is a natural number, and in the case of n−m <1, it is defined as the distance to T (n−m + N)).

  FIG. 44 (b) is a flow chart showing the procedure of end point reconstruction based on such four-point end point removal processing. First, in step SS11, the parameter n indicating the number of the normal end point to be subjected to the removal determination is set to n = 1, and in step SS12, it is determined whether L2 ≦ Wmerge. If an affirmative determination is made here, it is determined in step SS13 whether four points T (n-1), T (n), T (n + 1) and T (n + 2) lie on a straight line. It is judged. Here, if a positive determination is made, one of T (n) and T (n + 1) will be the removal target, but in order to decide which should be the removal target, in step SS14, A condition determination of L1 ≦ L3 is performed (in place of this, a condition determination of L1 <L3 may be performed). If an affirmative determination is made here, end point T (n) is to be removed in step SS15, and if a negative determination is made, end point T (n + 1) is to be removed in step SS16. .

  On the other hand, if a negative determination is made in step SS13, it is determined in step SS17 whether the three points T (n-1), T (n) and T (n + 1) are on a straight line. Here, if a positive determination is made, the process proceeds to step SS15, the end point T (n) is targeted for removal, and if a negative determination is made, the process proceeds to step SS18, T (n), T (n + 1), It is determined whether the three points of T (n + 2) lie on a straight line. Then, if an affirmative determination is made in step SS18, the process proceeds to step SS16, and the end point T (n + 1) is set as a removal target. This is the same process as the three-point end point removal process described above. If a negative determination is made in step SS12 or if a negative determination is made in step SS18, the process proceeds to step SS19 and processing in which neither T (n) nor T (n + 1) is to be removed Is done.

  Such processing is repeatedly performed through steps SS20 and SS21 while incrementing the parameter n by one until n ≧ N is satisfied. If the normal end point T (n) has already been removed, the processing procedure for the parameter n can be skipped. Then, when n ≧ N is satisfied, the process proceeds from step SS20 to SS22, and the normal end points to be removed are collectively removed.

  In the above processing procedure, in the case of n = 1, the removal determination target on the left side in FIG. 44 is the first normal end point T (1), so the normal end point T (n−1) in each of the above determination conditions ) Should be read as the N-th normal end point T (N). Similarly, in the case of n = N-1, the removal determination target on the left is the (N-1) th normal end point T (N-1), and the removal determination target on the right is the Nth normal end point T (N) As this is the case, the normal end point T (n + 2) in each of the above determination conditions needs to be read as the first normal end point T (1). Further, in the case of n = N, since the removal determination target on the left side is the N-th normal end point T (N), the normal end point T (n + 1) in each of the above determination conditions is the first normal end point T (N). It is necessary to read 1), and the end point T (n + 2) usually needs to be read as the second normal end point T (2).

After all, the end point reconstructing unit 116 can calculate the total N defined for one contour line of one unit figure, as a general theory, when adopting the reconstruction by the four-point end point removal processing described here as a general theory. The normal end points are referred to as first normal end point T (1) to Nth normal end point T (N) in the order along the contour line, and n-th normal end point T (n ) And a normal end point T (n-m) (where m is a natural number, and T (n-m + N) if n-m Assuming that the distance is L1 and the n-th normal end point T (n) and the (n + 1) -th normal end point T (n + 1) (where n = N, the first normal end point T (1): the same applies hereinafter) Between the (n + 1) th normal end point T (n + 1) and the (n + 2) th normal end point T (n + 2). However, when the distance between the first normal end point T (1) in the case of n = N-1 and the second normal end point T (2) in the case of n = N is the same as L3 is L3. ,
L2 is less than or equal to a predetermined reference value Wmerge, and the normal end point T (n-1) (where n = 1, the Nth normal end point T (N): the same applies hereinafter), T (n), T When the four points (n + 1) and T (n + 2) lie on a straight line, the normal end point T (n) is targeted for removal if L1 <L3, and the normal end point T (n + 1) if L1> L3. If L1 = L3, then either normal endpoint T (n) or normal endpoint T (n + 1) is targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n-1), T (n) and T (n + 1) are on a straight line, but the normal end point T (n + 2) is If it is not on the line, the end point T (n) is usually targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n), T (n + 1) and T (n + 2) are on a straight line, but the normal end point T (n-1) is a straight line If not on the line, the end point T (n + 1) is usually targeted for removal,
This process is repeated while increasing n by 1 until n ≧ N is satisfied, and the end point removal process is performed to remove the normal end point to be removed.

  Even when the four-point end point removal process is adopted as the end point reconstruction process performed by the end point reconstruction unit 116, almost the same result as that obtained when the three-point end point removal process described above is adopted can get. However, as described above, the end point removal process of the four-point system can select a more preferable end point as the removal target compared to the end point removal process of the three-point system. Can be made smaller.

(3) Reconstruction by Endpoint Relocation Processing In the 3-point or 4-point endpoint elimination processing described above, the endpoint reconstruction unit 116 reconstructs the endpoint by removing the existing ordinary endpoint. However, the end point repositioning process described here is an end point re-arrangement in which intermediate normal end points are once removed for a part of outlines, and new intermediate normal ends are arranged on each side of the part outlines. The arrangement process usually reconstructs the end point.

  FIG. 45 is a schematic diagram showing a procedure in the case where end point rearrangement processing is adopted as the end point reconstruction processing by the end point reconstruction unit 116 shown in FIG. Suppose now that endpoints A to H are defined on one side of a specific contour line of a specific unit figure, as shown in FIG. 45 (a typical endpoint is usually defined as an endpoint. (Defined by part 115). Here, for convenience of explanation, among these normal end points A to H, normal end points A and H constituting the vertex of a polygon to be a unit figure are referred to as corner normal points and normal end points B to G not constituting a vertex are indicated. We call it the intermediate normal endpoint.

  In the case of performing reconstruction by the end point rearrangement processing described here, the end point reconstruction unit 116 first performs processing for temporarily removing the intermediate normal end point defined on the contour line. FIG. 45 (b) shows a state in which intermediate normal end points B to G are removed from the normal end points A to H shown in FIG. 45 (a). Although only the upper side of the polygon forming one outline is shown in FIG. 45, actually, one outline forms a loop along each side of the polygon. For each edge of the outline, processing is performed to temporarily remove intermediate normal endpoints. FIG. 45 (b) shows a state in which only the corner normal end points A and H are left by removing the middle normal end point for the upper side of the outline.

  Subsequently, the end point reconstruction unit 116 performs end point rearrangement processing by repositioning the intermediate normal end points at predetermined intervals on each side of the outline from which the intermediate normal end points have been removed. FIGS. 45 (c), (d) and (e) show an example in which the middle normal end point is rearranged for the upper side of the above-mentioned outline. More specifically, FIG. 45 (c) is an example in which a predetermined reference interval Lreg is set, and the intermediate normal endpoints are rearranged at each reference interval Lreg starting from the position of the left end corner normal endpoint A as a starting point. . Normally, between the end points A to E, the interval between the adjacent end points is the reference interval Lreg, but the end point E to H is usually the fractional interval Lodd. For example, when the length of the side AH is 45 nm and the reference interval Lreg is set to 10 nm, the fractional interval Lodd is 5 nm.

  On the other hand, in FIG. 45 (d), between the normal end points C and D in the central portion is a fractional interval Lodd, and the normal end points A-B, B-C, D-E, and E-H are standard. It is an example set as the interval Lreg. Generally, in the case of a fine figure pattern such as a semiconductor device, the exposure error in the vicinity of the vertex of the polygon tends to be large, and the originally angular shape tends to be rounded. Under such circumstances, it is possible to reduce the computational load of the simulation if the space between the corner normal end point and the intermediate normal end point adjacent thereto is as far as possible. From such a point of view, the fractional interval Lodd, as shown in FIG. 45 (c), is located at the center of the side AH, as shown in FIG. 45 (d), rather than at a position adjacent to the corner normal end H. It is preferable to arrange near. When the length of the side AH is 45 nm, the intervals between the end points are 10, 10, 5, 10, and 10 nm in order from the left.

  FIG. 45 (e) is an example in which all the end point intervals are set at equal intervals Leven. If an optimal value is determined in advance as the value of Leven, the intermediate normal end point can be rearranged at a position obtained by equally dividing the side AH so as to obtain equally spaced Leven closest to the optimal value. For example, when the optimal value of Leven is set to 10 nm, assuming that the length of side AH is 45 nm, Leven = 11.25 nm when dividing it into four, Leven = 9 nm, 6 when dividing it into five In the case of division, Leven = 7.5 nm, so Leven = 9 nm is obtained as the value closest to the optimum value of 10 nm. Therefore, as illustrated, the side AH may be divided into five, and the intermediate normal end points may be rearranged at equal intervals Leven = 9 nm.

  FIG. 46 is a flow chart showing a basic procedure of reconstruction of the end point when the end point rearrangement process shown in FIG. 45 is adopted. First, in step SS31, a process of temporarily removing an intermediate normal end point (a normal end point not forming a vertex) on a specific contour line to be processed is performed. Subsequently, in step SS32, processing is performed to place new intermediate normal endpoints on each side of the polygon that constitutes the contour. A specific method is as shown in FIGS. 45 (c) to (e).

(4) Proper Use of End Point Removal Processing and End Point Relocation Processing So far, the three-point or four-point end point removal processing (FIGS. 41 to 44) and the end point relocation processing (FIGS. 45 to 46) have been described. Here, the end point removing process can be applied to any contour of any unit graphic, and may be applied to all contours of all unit graphics included in the original graphic pattern 10. On the other hand, there are cases where the end point relocation processing is inappropriate when applied depending on the outline.

  This is because, originally, the present invention is characterized by dividing a unit figure into a plurality of divided figures and defining an end point usually based on the vertex position of this divided figure, it is defined based on the vertex position of the divided figure Although the process of removing some normal endpoints (normally reducing the number of endpoints) does not lose the feature of "normal endpoints defined based on the vertex positions of the divided figure", This is because when the process of rearranging the normal end point (the process of correcting the position) is performed, the above-described feature is lost for the rearranged normal normal end point.

  For example, in the example shown in FIG. 42, the normal end points D and I shown by white triangles are removed, but each other normal end point is “four sets of divided figures 10a shown in FIG. It has a feature of “normal end point defined based on vertex positions of ̃10 d. Here, if each normal end point is defined at a position based on the vertex position of divided figures 10a to 10d, as described in FIGS. 36 and 37, in the exposure process based on corrected figure pattern 15, the shot of the exposure beam The advantage is that the number can be reduced. This is because the horizontal positions of the intermediate normal endpoints B, C and D defined on the upper side of the unit graphic 10 shown in FIG. 38 (d) and the horizontal positions of the intermediate normal endpoints I, H and G defined on the lower side. In order to be aligned.

  However, when the end point rearranging process is performed on the example of the unit graphic 10 shown in FIG. 38 (d), the positions of the corner normal end points A, E, F, J do not change, but the intermediate normal end points B, C, The positions of D, G, H, and I are corrected and rearranged at positions unrelated to the vertex positions of the four sets of divided figures 10a to 10d. In such a case, the horizontal position of the intermediate normal end defined in the upper side of the unit graphic 10 may not be aligned with the horizontal position of the intermediate normal end defined in the lower side, which is advantageous in that the number of exposure beam shots can be reduced. It may not be possible.

  Therefore, with regard to the unit graphic shown in FIG. 42 and the unit graphic shown in FIG. 43, it is preferable to rearrange the end points based on the end point removing process (FIGS. 41 to 44). On the other hand, with respect to a U-shaped unit figure (so-called unit figure of hole opening) as shown in FIG. 40 (f), end point removal processing for the outer outline Oout (FIGS. 41 to 44) It is preferable to rearrange the end points based on the above, and to rearrange the end points based on the end point rearrangement process (FIGS. 45 to 46) for the inner contour Oin.

  FIG. 47 shows the end point rearrangement process according to the procedure of FIG. 46 applied to the normal end points MT on the inner contour Oin of the unit-shaped figure 10 shown in FIG. 40 (f). It is a top view which shows the example which performed reconstruction of. Specifically, intermediate normal end points O and P on the lower side NQ of the rectangle forming the inner contour Oin and intermediate normal end points S and T on the upper side MR are once removed, and a new intermediate normal is deleted on the lower side NQ. An end point U is disposed, and a new intermediate normal end point V is disposed on the upper side MR.

  By performing such end point rearrangement processing, intermediate normal end points U and V on the inner contour Oin are at positions irrelevant to the vertex positions of the eight sets of divided figures 10a to 10h shown in FIG. 40 (b). It will be arranged. However, since the inside of the figure surrounded by the outer outline Oout becomes the exposure area, the inside of the figure surrounded by the inside outline Oin becomes the unexposed area, so the intermediate normal end point U on the inside outline Oin Even if the positions of V and V are set to arbitrary positions, the merit of reducing the number of shots of the exposure beam does not affect.

  On the other hand, intermediate normal end points B, C, E, F, H, I, K, L on the outer contour Oout must be sure as shown in the arrangement of the respective intermediate normal end points shown in FIG. 47 (a). Corner normal points A, D, G, and J are defined at positions somewhat away from the basic points (basically, positions separated by Wsplit), while intermediate normal points O, P, and C on the inner contour Oin S, T may be defined in the immediate vicinity of the corner normal end points M, N, Q, R. Moreover, as described above, the exposure error in the vicinity of the apex of the polygon becomes large, and the originally angular shape tends to be rounded. Therefore, if the distance between the corner normal end point and the intermediate normal end point adjacent thereto is as large as possible, the computational load on the simulation can be reduced.

  Under such circumstances, the end points are rearranged based on the end point removing process (FIGS. 41 to 44) for the intermediate normal end points defined on the outer contour Oout, according to the vertex positions of the divided figures. It is preferable that the arrangement be maintained as it is, so that the merit that the number of shots of the exposure beam can be reduced can be enjoyed. On the other hand, for the intermediate normal end points defined on the inner contour Oin, the end points are rearranged based on the end point rearrangement process (FIGS. 45 to 46), and the end points are separated as far as possible from the corner normal end points. It is preferable to arrange new intermediate normal endpoints so that the advantage of reducing the computational load of simulation can be enjoyed.

  As described in 6.2 6.2, with regard to the unit figure 10 of the hole opening, the figure that constitutes the outer outline Oout is called "a positive figure", and the figure that constitutes the inner outline Oin is called a "anti-figure". Then, when the unit graphic vector group makes a round in the forward direction (clockwise direction in the case of the embodiment described herein) along the contour, the contour is a positive figure constituting the outer contour Oout of the unit graphic 10. If it can be determined that the unit graphic vector group makes a round along the contour in the opposite direction (counterclockwise in the case of the embodiment described herein), the contour is the inner contour Oin of the unit graphic 10. It can be judged that it is an anti-figure that constitutes

  Therefore, when there is a possibility of handling a unit figure of a hole, the end point restructuring unit 116 performs an end point removing process (FIGS. 41 to 44) and an end point rearrangement process (FIGS. 45 to 46) as a unit figure. In the case of the example shown in FIG. 47 (a), the outer contour Oout is provided with a function to selectively execute each of the contours in the forward direction. For the normal end points above, the end point removal processing is performed to perform end point reconstruction, and the normal end points on the opposite figure surrounded by unit graphic vectors around the reverse direction opposite to the forward direction (see FIG. 47 (a) In the case of the example shown, end point reconstruction may be performed by end point rearrangement processing for the normal end point on the inner contour Oin.

<6.4 Definition of evaluation points>
Here, the function of the evaluation point definition unit 117 in the evaluation point setting unit 110 shown in FIG. 32 will be described. The evaluation point definition unit 117 is usually based on the normal end point defined on the outline of the unit graphic by the end point definition unit 115 or in the case of the embodiment in which the end point reconstruction unit 116 described in 6.3 6.3 is provided. This is a component that defines an evaluation point E on the basis of a normal end point obtained by the end point reconstruction unit 116. That is, the evaluation point definition unit 117 performs processing of defining an evaluation point between two normal end points arranged adjacent to each other on the contour line of each unit graphic.

  The position defining each evaluation point E may be any position in principle as long as it is between adjacent pairs of normal endpoints, but in practice, the midpoint position of a pair of normal endpoints It is preferable to define in. This is because it is more appropriate to set the evaluation point at the center point of the contour line segment when calculating the amount of deviation (process bias) for the contour line segment sandwiched by a pair of normal end points by simulation. It is considered to be obtained.

  Therefore, in the embodiment described here, the evaluation point definition unit 117 performs processing of defining evaluation points at middle point positions of two normal end points disposed adjacent to each other on the contour line of the unit graphic. FIG. 48 (a) shows a normal end point (x mark) on the parallelogram type unit graphic 10 after the end point reconstruction shown in FIG. 42 (b) (each end point is relabeled with A to H. FIG. 48 (b) shows an example in which evaluation points 1 to 8 (black circles) are defined based on these normal end points A to H. FIG. Specifically, on the contour line segment AB sandwiched between a pair of normal end points AB, an evaluation point 1 is defined at the center position thereof, and a contour line segment BC sandwiched between a pair of normal end points B-C. On the top, evaluation point 2 is defined at the center position,..., Evaluation point 8 is defined at the center position on the contour line segment HA sandwiched between a pair of normal end points HA .

  Similarly, FIG. 49 (a) shows a normal end point (x mark) on the L-shaped unit graphic 10 after the end point reconstruction shown in FIG. 43 (b) (each end point has a symbol A to H). FIG. 49 (b) shows an example in which evaluation points 1 to 8 (black circles) are defined based on these normal end points A to H. FIG. Each evaluation point is defined at the midpoint of each contour line segment. Also, FIG. 50 (a) shows a normal end point (marked with x) on the B-shaped unit graphic 10 after the end point reconstruction shown in FIG. 47 (b) (each end point has a symbol A to R FIG. 50 (b) shows an example in which evaluation points 1 to 18 (black circles) are defined based on these normal end points A to R). Again, each evaluation point is defined at the midpoint of each contour line segment.

  The evaluation points thus defined correspond to the evaluation points E11, E12, and E13 shown in FIG. 3, and the feature amounts x1 to xn for these evaluation points are determined by the feature amount extraction unit 120 shown in FIG. The extracted value y is estimated by the bias estimation unit 130 for the process bias for each of these evaluation points. Then, based on the estimated value y of the process bias obtained for each evaluation point, the pattern correction unit 140 corrects the position of the contour line segment including the evaluation point.

<6.5 Definition of replenishment end point>
Subsequently, the function of the replenishment end point definition unit 118 in the evaluation point setting unit 110 shown in FIG. 32 will be described. The replenishment end point definition unit 118 searches an intermediate normal end point, which is a normal end point that does not constitute a polygon vertex, for each unit graphic, and additionally defines a replenishment end point at the same position as the searched intermediate normal end point. Play. As described above, the normal end points defined by the normal end point definition unit 115 and the normal end points obtained by reconstruction by the end point reconstruction unit 116 are the corner normal end points constituting the vertices of the polygon and the vertices of the polygon. It is divided into intermediate normal endpoints that are not configured. The replenishment end point definition unit 118 additionally defines the replenishment end point at the same position as the intermediate normal end point.

  Here, first, in addition to the normal end points, the necessity of additionally defining the replenishment end point will be described. As described above, when the evaluation point definition processing by the evaluation point definition unit 117 is completed, an evaluation point is determined at the midpoint position for each of the contour line segments sandwiched between a pair of normal end points adjacent to each other. For example, in the case of the example shown in FIG. 48 (b), an evaluation point 1 is defined on a contour line segment AB sandwiched between a pair of normal end points AB and a contour line sandwiched between a pair of normal end points BC A score of 2 is defined on the minute BC. Then, a feature amount is extracted for each evaluation point, an estimated value y of the process bias is determined, and the position of the contour line segment including the evaluation point is corrected based on the estimated value y of the process bias.

  For example, in the example shown in FIG. 48 (b), as a result of simulation, an estimated value y (1) =-1 nm of process bias is obtained for evaluation point 1, and an estimated value y of process bias is obtained for evaluation point 2. Let) = -2 nm be determined. In this case, when the lithography process is performed using the parallel figure-shaped original figure pattern 10 shown in FIG. 48B as it is, the point corresponding to the evaluation point 1 on the actual figure pattern 20 obtained on the actual substrate S is It is estimated by simulation that the point shifted to the inside by 1 nm and the point corresponding to the evaluation point 2 slipped to the inside by 2 nm. Therefore, the pattern correction unit 140 moves the contour line segment AB having the evaluation point 1 outward by 1 nm, and moves the contour line segment BC having the evaluation point 2 outside by 2 nm, for example. Correction processing will be performed.

  At this time, moving the contour line segment AB and the contour line segment BC independently can be easily understood as a geometric concept. However, as data processing on the computer, it will be disadvantageous if the intermediate normal end point B remains at a single end point. In the present invention, as shown in the example of FIG. 3B, correction is performed to move each evaluation point in the direction orthogonal to the contour. Therefore, in order to move the contour line segment AB and the contour line segment BC in accordance with the respective correction amounts, the normal end point B which serves as the right end of the contour line segment AB and the left end of the contour line segment BC The normal end point B, which plays the role of, needs to be treated as a separate end point for data processing. From such circumstances of computer processing, in the present invention, the replenishment end point B 'is additionally defined at the same position with respect to the intermediate normal end point B. Thus, the replenishment end point B 'additionally defined can be treated as an end point of only one contour line segment, and the existing middle normal end point B can be treated as an end point of only the other contour line segment.

  For example, in the case of a parallelogram-shaped unit figure shown in FIG. 48 (b), replenishment end points B ', C', F 'and G' are additionally defined at the same positions for intermediate normal end points B, C, F and G, respectively. Then, as shown in FIG. 51 (a), in addition to the eight normal end points A to H (x marks), four additional replenishment end points B ', C', F ', G' (square marks) Is added, and a total of 12 end points will be defined. Then, for example, the left end of contour line segment AB is normal end point A, the right end is replenishment end point B ', the left end of outline line segment BC is normal end point B, the right end is replenishment end point C', and the left end of outline line segment CD Is the normal end point C, the right end is the normal end point D, the upper end of the contour line segment DE is the normal end point D, the lower end is the normal end point E, the right end of the outline line segment EF is the normal end point E, the left end is the replenishment end point F ' The right end of line segment FG is normal end point F, the left end is replenishment end point G ', the right end of outline line segment GH is normal end point G, the left end is normal end point H, and the lower end of outline line segment HA is normal end point H, upper end is normal It can be an end point A.

  By defining both ends of eight sets of contour line segments in this manner, for example, contour line segment AB and contour line segment BC can be moved independently (for computer processing, normal end point B and replenishment end point) B 'can be moved to another coordinate position). The reason why the additional definition of the replenishment end point is made only for the intermediate normal end point is that one intermediate normal end point may be separated into two end points occupying two different positions after correction as described above. In contrast, one corner normal end point remains as one corner normal end point even after correction. Such circumstances will be described in な が ら 7 with specific examples.

  Similarly, in the case of the L-shaped unit graphic shown in FIG. 49 (b), if the replenishment end points F 'and H' are additionally defined at the same positions for the intermediate normal end points F and H, respectively, in FIG. As shown, in addition to the eight normal end points A to H (x marks), two additional replenishment end points F 'and H' (square marks) are added, and a total of 10 end points are defined. Become. Then, for example, the right end of contour line segment EF can be normal end point E, the left end can be replenishment end point F ', the right end of outline line segment FG can be normal end point F, and the left end is normal end point G. The lower end of line segment GH can be normal end point G, the upper end can be replenishment end point H ', the lower end of contour line segment HA can be normal end point H, and the upper end can be normal end point A. Processing is possible.

  Further, in the case of a square-shaped unit graphic shown in FIG. 50 (b), the replenishment end point B is placed at the same position for each of the intermediate normal end points B, C, E, F, H, I, K, L, O, R. If ', C', E ', F', H ', I', K ', L', O 'and R' are additionally defined, 18 normal endpoints A, as shown in FIG. In addition to ~ R (x mark), 10 additional replenishment end points B ', C', E ', F', H ', I', K ', L', O ', R' (square marks) A total of 28 end points will be defined, being added. Then, for example, the left end of contour line segment AB is normal end point A, the right end is replenishment end point B ', the left end of outline line segment BC is normal end point B, the right end is replenishment end point C', and the left end of outline line segment CD Can be set as the end point C, the right end as the end point D, and so on, and processing for independently moving each contour line segment becomes possible.

  FIG. 52 is a schematic diagram (figures (a), (b)) and a flow chart (figure (c)) showing the specific procedure of the replenishment end point definition process performed by the replenishment end point definition unit 118 shown in FIG. As described above, the replenishment end point definition unit 118 searches for an intermediate normal end point for each unit graphic, and additionally defines a replenishment end point at the same position as the searched intermediate normal end point. Here, the search process of the intermediate normal end point can be specifically performed in the following procedure.

  Now, it is assumed that a total of N normal endpoints are defined on a specific contour line of a specific unit graphic to be processed. Among these N normal end points, a part is a corner normal end point constituting a vertex of a polygon, and a remaining part is an intermediate normal end point not constituting a vertex of a polygon. The search process performed here is a process of finding an intermediate normal end point out of N normal end points.

  Therefore, as shown in FIG. 52 (a), all the N normal end points are set as the first normal end point T (1) to the N-th normal end point T (N) in the order along the outline. It is determined whether the intermediate normal end point is the intermediate normal end point with the n-th normal end point T (n) as the determination target. As shown in the figure, three (n-1) normal end points T (n-1), n-th normal end points T (n), and (n + 1) normal end points T (n + 1) are used for this determination. Use. Oa in the figure is a contour segment connecting two end points T (n-1) and T (n), and Ob in the diagram is a contour segment connecting two end points T (n) and T (n + 1). is there. When the n-th normal end point T (n) is determined to be an intermediate normal end point, the replenishment end point T is located at the same position as the n-th normal end point T (n) as shown in FIG. 52 (b). Define (n) 'additionally. In the figure, normal end points T (n) are indicated by x marks, and replenishment end points T (n) 'are indicated by square marks.

  If the replenishment end point T (n) 'is additionally defined at the same position when the n-th normal end point T (n) is the intermediate normal end point, the left end of the contour line segment Oa is usually the end point T (n-1) And the right end can be the replenishment end point T (n) '. Further, the left end of the contour line segment Ob can be set as the normal end point T (n), and the right end can be set as the normal end point T (n + 1). Since the end points T (n) and T (n) ′ can be moved independently and independently, the contour line segments Oa and Ob can be moved independently and the pattern correction unit 140 There is no problem in performing the correction process.

  FIG. 52 (c) is a flowchart showing a specific procedure of replenishment end point definition processing performed by the replenishment end point definition unit 118. First, in step SS41, a parameter n indicating the number of the normal end point to be determined whether or not it is an intermediate normal end point is set to n = 1, and in step SS42, the (n-1) th normal end point T It is determined whether or not three points of n-1), n-th normal end point T (n) and (n + 1) -th normal end point T (n + 1) are on a straight line. Here, if a positive determination is made, the end point T (n) to be determined is an intermediate normal end point (end point that does not constitute the vertex of the polygon), so in step SS43 this intermediate normal end point T At the same position as (n), a replenishment end point T (n) 'is additionally defined. If a negative determination is made in step SS42, the end point T (n) to be determined is the corner normal end point (the end point constituting the vertex of the polygon), so the replenishment end point T (n) ' There is no additional definition.

  Such processing is repeatedly performed until n = N is reached through steps SS44 and SS45 while the parameter n is incremented by one. Then, when n = N is reached, the replenishment end point definition process is completed. By performing such processing, replenishment end points indicated by square marks in FIGS. 51 (a), (b) and (c) are defined.

  In general terms, the replenishment end point definition unit 118 determines the total of N normal end points defined for one outline, in the order of the first outline from the first normal end point T (1) to the first When the normal end point T (N) of N is called, noting the n-th normal end point T (n), the (n-1) th normal end point T (n-1) (where n = 1) In the case, the Nth normal end point T (N), the nth normal end point T (n), and the (n + 1) th normal end point T (n + 1) (however, in the case of n = N, the first normal end point T (n If the three points in 1) are on a straight line, the process of determining the nth normal end point T (n) as an intermediate normal end point and additionally defining a replenishment end point at the same position as the intermediate normal end point is n = It may be repeatedly executed from 1 to N.

<<< 7. 7. Details of pattern correction unit >>>
Subsequently, the pattern correction unit 140 will be described in detail. FIG. 53 is a block diagram showing a pattern correction unit 140 in a figure pattern shape correction apparatus 100 according to the basic embodiment of the present invention shown in FIG. 1 and components related thereto. The basic function of the pattern correction unit 140 is to correct the position of each evaluation point E set by the evaluation point setting unit 110 based on the estimated value y of the process bias output from the bias estimation unit 130, to thereby obtain an original figure pattern. Correction for 10 is performed to create a corrected figure pattern 15.

  Here, the function of the evaluation point setting unit 110 is as described in detail in §6, and processing for defining the normal end point T, the replenishment end point T ′ and the evaluation point E for each unit figure constituting the original figure pattern 10 Is done. The information of each of these points can be said to be information of the contour line segment O. That is, the position information of each normal end point T and each replenishment end point T ′ is information indicating the positions of both ends of a specific contour line segment O, and the position information of the evaluation point E is on the contour line segment O It is information indicating the position of the defined evaluation point. The pattern correction unit 140 creates the corrected figure pattern 15 by performing movement processing and expansion processing on the contour line segment O and addition processing of a new contour line segment based on the information on the contour line segment O.

  As shown in FIG. 53, the pattern correction unit 140 has a contour line segment movement processor 141, a contour line segment addition processor 142, and a contour line segment expansion and contraction processor 143. The functions of these components will be specifically described below by taking an L-shaped unit graphic shown in FIG. 51 (b) as an example. As described above, the figure pattern shape correction apparatus 100 according to the present invention is actually configured by incorporating a dedicated program into a computer. Therefore, the above-described components 141, 142, and 143 included in the pattern correction unit 140 are actually configured by incorporating a dedicated program into the computer.

<7.1 Contour line segment movement processing>
Here, first, the function of the contour line segment movement processing unit 141 shown in FIG. 53 will be described. As illustrated, the contour line segment movement processing unit 141 includes information on the contour line segment O defined by the evaluation point setting unit 110 together with the information on the original figure pattern 10 (normal end point T, replenishment end point T ′, evaluation point E Position coordinates and information indicating their interrelationship). Also, from the bias estimation unit 130, an estimated value y of the process bias is given for each evaluation point E.

  FIG. 54 is an enlarged plan view showing each end point and each evaluation point defined on the L-shaped unit graphic shown in FIG. 51 (b). In this figure, each normal end point A to H is indicated by an x mark, each replenishment end point F ′, H ′ is indicated by a square mark, each evaluation point 1 to 8 is indicated by a circle, and reference numerals 1 to 8 of each evaluation point Is described in the circle. Moreover, about the outline line segment pinched | interposed by each adjacent end point, the code | symbol of O1-O8 is attached | subjected and shown. For example, a portion sandwiched by the end points A-B is a contour line segment O1, and a portion sandwiched by the end points B-C is a contour line segment O2. Of the eight normal end points A to H, six normal end points A, B, C, D, E and G are corner normal end points constituting the vertex of the polygon, and the remaining two normal end points F and H are intermediate ordinary endpoints that do not constitute the vertices of the polygon. As described above, replenishment end points F 'and H' are respectively defined at the same position for the two intermediate normal end points F and H.

  FIG. 55 is a diagram showing an example of a data format representing the mutual relationship between each end point and each evaluation point shown in FIG. As described above, the information of the contour line segment O is given from the evaluation point setting unit 110 to the contour line segment movement processing unit 141, and the information of the contour line segment O is specifically shown in FIG. It will be given in the data format shown. In this data format, one end point and one evaluation point correspond to a data pair (enclosed in parentheses in the figure), and these data pairs are arranged along the contour of the unit figure. They are arranged in the order of clockwise.

  For example, the first data pair (A, 1) is a data pair in which the end point A and the evaluation point 1 are associated, and evaluation is performed on the contour line segment O1 starting at the end point A and ending at the end point B It indicates that point 1 is defined. Similarly, the second data pair (B, 2) is a data pair in which the end point B and the evaluation point 2 are associated with each other, on the contour line segment O2 starting at the end point B and ending at the end point C. It shows that the evaluation point 2 is defined.

  The sixth data pair (F ′,)) is a data pair in which the replenishment end point F ′ and the evaluation point 対 応 are associated with each other. Here, the evaluation point Φ is an evaluation point that does not actually exist, and the data “Φ” is empty data provided for the convenience of expressing the replenishment end point F ′ in the form of a data pair. Similarly, the ninth data pair (H ′,)) is also a data pair in which the replenishment end point H ′ and the evaluation point 対 応 are associated, and the data “Φ” is null data.

  If the data format shown in FIG. 55 is adopted, it is possible to represent all the end points, which normally wrap the end points and the replenishment end points, as data pairs. Of course, since the data constituting each data pair includes position coordinate values of each end point and each evaluation point, the position of each end point or each evaluation point on a two-dimensional plane can be specified. Moreover, since each data pair is arranged in the clockwise order along the contour of the unit figure, it is possible to recognize the contour line segment sandwiched between the adjacent end points, and it is defined on the contour line segment Evaluation points can be recognized. For example, the position of contour line segment O1 sandwiched between adjacent end points AB can be specified by two data pairs of (A, 1) and (B, 2), and the position on contour line segment O1 can be specified. The position of the evaluation point 1 of can be specified.

  In the case of the unit figure of a hole as shown in FIG. 51 (c), a regular figure forming the outer contour Oout and an opposite figure forming the inner contour Oin are formed, but the upper figure is the upper one. As in the example, it is preferable to list each data pair in the clockwise order along the outline, and for the anti-figure, to arrange each data pair in the counterclockwise order along the outline contrary to the above example . Then, it becomes possible to identify whether each outline is the outer outline Oout or the inner outline Oin based on the order of arrangement of each data pair, and when performing movement processing of outline segments described later To the inside and outside of the unit figure.

  In the embodiment shown in FIG. 53, the original figure pattern 10 is given to the contour line segment movement processing unit 141, but the contour line segments from the evaluation point setting unit 110 in performing the contour line segment movement processing It is sufficient if the data shown in FIG. 55 is given to the movement processing unit 141 and the estimated value y of the process bias for each evaluation point is given from the bias estimation unit 130 to the contour line segment movement processing unit 141. That is, the original figure pattern 10 is not essential information for the contour line segment movement processing unit 141. However, since the original figure pattern 10 is information directly indicating the shape of the unit figure, in practical use, it is given to the outline line segment movement processing unit 141 for reference when performing outline line segment movement processing. It is preferable to provide.

  FIG. 56 is a plan view showing a state in which the contour line segment movement processing unit 141 has performed contour line segment movement processing on the L-shaped unit graphic 10 shown in FIG. The outline line segment shown by a broken line in the figure shows the state before the movement process, and the outline line segment shown by the solid line in the figure shows the state after the movement process. The movement processing of the contour line segment is such that, for each evaluation point, a normal end point or a replenishment end point adjacent to one side of the evaluation point along the contour line of the unit figure is a first end point and is adjacent to the other side. A contour line segment whose normal end point or replenishment end point is the second end point is defined, and the contour line segment for each evaluation point is a contour line segment by a correction amount according to the estimated value of the process bias of the evaluation point. It is performed by moving in the orthogonal direction. At this time, either the normal end point or the replenishment end point is selected for each contour line segment so that each end point of one contour line segment is not used redundantly as each end point of another contour line segment.

  The example shown in FIG. 56 is an example in which the estimated value y of the process bias of each of the evaluation points 1 to 8 takes a negative value (the outline of the original unit figure 10 is shifted inward by simulation of the lithography process) It shows the correction process for the example in which the result is obtained. That is, when the lithography process is performed using the original unit figure 10 as it is, the actual figure 20 which is smaller than the original unit figure 10 is obtained. Therefore, the original unit figure 10 is slightly enlarged and corrected. A correction process for obtaining a unit graphic 15 is performed.

  Bold arrows shown in FIG. 56 indicate the moving direction of each evaluation point. In the example shown here, since the estimated value y of the process bias of the evaluation points 1 to 8 takes a negative value, all of the evaluation points 1 to 8 are moved in the outward direction of the unit graphic. For example, when the estimated value of the process bias at the evaluation point 1 is y (1) =-1 nm, the evaluation point 1 is moved outward by 1 nm as indicated by the thick arrow in the figure. As described above, in the present invention, the movement of the evaluation point is performed in the direction orthogonal to the contour line segment including the evaluation point. In addition, the evaluation point is moved for each contour line segment including the evaluation point. Therefore, in the case of the illustrated example, the evaluation point 1 is moved outward by 1 nm as indicated by a thick arrow, and the contour line segment O1 including the evaluation point 1 is also moved in parallel in the same manner.

  Similarly, in the example shown in FIG. 56, the outline line segments O2 to O8 are translated outward. The point to be noted here is the behavior of the intermediate normal end points F and H and the replenishment end points F 'and H' additionally defined at the same positions as these. In the illustrated example, since the estimated value y of the process bias for the evaluation point 5 and the estimated value y of the process bias for the evaluation point 6 are different values, the movement amount of the outline line segment O5 and the outline line segment The moving amount of O6 also takes different values. For this reason, since the movement amounts of the contour line segments O5 and O6 are different even if the movement direction is the same, as shown in the drawing, both are separated after the movement processing. However, since both ends of the contour line segment O5 are end points E and F 'and both ends of the contour line segment O6 are end points F and G, even if the positions of the end points F and F' are different, There is no problem.

  The same applies to the contour line segments O7 and O8. As illustrated, since the contour line segment O7 and the contour line segment O8 have different amounts of movement, both are separated after the movement processing. However, since both ends of the contour line segment 7 are end points G and H 'and both ends of the contour line segment 8 are end points H and A, even if the positions of the end points H and H' are different, There is no problem. That is, the substantial processing on the computer in this movement processing is the correction of the data shown in FIG. 55, but there is no need to add any correction to the data format structure itself shown in FIG. And it only needs to perform processing to correct the coordinate values of each evaluation point. The replenishment end point definition unit 118 additionally defines the replenishment end points F ′ and H ′ for the intermediate normal end points F and H in order to facilitate such convenience.

  On the other hand, no additional definition of the replenishment end point is made for the corner normal end points A, B, C, D, E and G. Therefore, these corner normal end points are separated into two different positions after the movement processing, as shown in the figure. For example, the corner normal end point B is separated into the position of the right end of the contour line segment O1 and the upper end position of the contour line segment O2 after the movement processing. The handling of the corner normal end point separated into two after the movement processing in this way is performed by the outline line segment expansion and contraction processing unit 143 described later.

<7.2 Contour Line Addition Processing>
Subsequently, the function of the contour line segment addition processing unit 142 illustrated in FIG. 53 will be described. When the intermediate normal end point and the replenishment end point, which were at the same position as each other before the movement processing, are separated into two different positions after the movement processing, the contour line segment addition processing unit 142 connects the positions after the movement processing. It has a function to perform additional processing to add a new contour line segment.

  In the case of the example shown in FIG. 56, the intermediate normal end points F and H and the replenishment end points F ′ and H ′ are arranged at different positions by the movement processing of the contour line segment. Therefore, a gap is generated between the intermediate normal end point F and the replenishment end point F ', and a gap is also generated between the intermediate normal end point H and the replenishment end point H'. The contour line segment addition processing unit 142 performs a process of adding a new contour line segment to fill the gap.

  FIG. 57 is a plan view showing a state in which the contour line segment addition processing unit 142 has performed contour line segment addition processing on the unit graphic after the movement processing shown in FIG. The portions shown by thick lines in the figure are newly added contour line segments. Specifically, for the intermediate normal end point F and the replenishment end point F 'that were at the same position as each other before the movement processing, a new outline line segment O5' connecting the positions after the movement processing is added, and before the movement processing A new contour line segment O7 'is added to connect the positions after movement processing for the intermediate normal end point H and the replenishment end point H' that are at the same position as each other.

  However, in the case of the embodiment described here, as described above, since the information of each end point and each evaluation point constituting the unit graphic is expressed in the data format illustrated in FIG. The contour line segment addition process is also automatically performed by performing the process of correcting the coordinate values of the end point and the evaluation point in each data pair indicated by the above data format (the contour line segment moving process described above).

  For example, the sixth data pair (F ',)) in FIG. 55 is data corresponding to the contour line segment O5' newly added in FIG. Specifically, data “F ′” in data pair (F ′,)) is information indicating the coordinate value of end point F ′ which is the beginning of contour segment O5 ′, and data “Φ” is contour segment O5. It is information which shows the position of the evaluation point about '. Actually, since the data "「 "is null data, the contour line segment O5 'is a contour line segment having no evaluation point. The coordinate value of the end point F which is the end of the contour line segment O5 'is indicated by data "F" in the seventh data pair (F, 6).

  Similarly, the ninth data pair (H ',)) in FIG. 55 is data corresponding to the contour line segment O7' newly added in FIG. Specifically, data "H '" in data pair (H',)) is information indicating the coordinate value of end point H 'which is the beginning of contour line segment O7', and data "Φ" is contour line segment O7. It is information which shows the position of the evaluation point about '. Actually, since the data "「 "is null data, the contour line segment O7 'is a contour line segment having no evaluation point. The coordinate value of the end point H which is the end of the contour line segment O7 'is indicated by the data "H" in the tenth data pair (H, 8).

  As a result, when the data format illustrated in FIG. 55 is used, the movement processing by the contour line segment movement processing unit 141 automatically performs the addition processing by the contour line segment addition processing unit 142. That is, end points F and F 'and end points H and H' defined at the same position before movement processing move to different positions, and in each data pair shown in FIG. 55, these end points F, F ', H and H When the coordinate value of 'is rewritten to the new coordinate value after movement processing, a new outline line segment O5' connecting the end points F and F 'is automatically added, and the end points H and H' are connected. A new contour line segment O7 'is automatically added. Therefore, when the data format illustrated in FIG. 55 is used, the addition processing by the contour line segment addition processing unit 142 is realized by the movement processing by the contour line segment movement processing unit 141, and in data processing, the addition processing is newly performed. There is no need to perform various data processing tasks.

<7.3 Contour Line Segment Expansion Processing>
Next, the function of the contour line segment expansion and contraction processing unit 143 shown in FIG. 53 will be described. As described in 7.1 7.1, when moving processing is performed, the corner normal end point is separated into two positions. For example, in the case of the example shown in FIG. 57, corner corner end points A (shown as A0, A1 and A2 in the figure) and B, C, D, E and G have two positions after moving processing. It is separated. The contour line segment expansion / contraction processing described here aims at creating a new corner normal end point by fusing two corner normal ends separated in this way.

  Therefore, when the corner normal end point, which is a normal end point that has been a vertex of the polygon before the movement processing, is separated into two different positions after the movement processing, the contour line segment expansion / contraction unit 143 An expansion and contraction process is performed to expand and contract each outline line segment including the corner normal end point after separation so that the two corner normal ends merge into a new corner normal end point.

  Specifically, in the case of the embodiment described here, the contour line segment expansion / contraction unit 143 separates one corner normal end point into a first moving end point and a second moving end point by moving processing. An intersection point of a straight line including a first post-moving contour line segment having the first moving end point as one end point and a straight line including a second post-moving contour line segment having the second moving end point as one end point; The extension process is performed such that the intersection point is a new corner normal end point, and one end point becomes a new corner normal end point with respect to the first post-movement contour line segment and the second post-movement contour line segment Do.

  Here, a specific procedure of the expansion and contraction process will be described by focusing on the corner normal end point A in the example shown in FIG. Here, as shown in FIG. 57, the corner normal end point A before movement processing is indicated by the symbol A0, and the corner normal end point A0 corresponds to the first movement end point A1 and the second movement end point A2 after movement processing. Consider the case of separation. In this case, a straight line (indicated by a dot-and-dash line in the figure) including a first post-moving contour line segment O8 having the first moving end point A1 as one end point and a second using the second moving end point A2 as one end point An intersection point J with a straight line (indicated by an alternate long and short dash line in the figure) including the post-movement outline line segment O1 is determined. Then, this intersection point J is set as a new corner normal end point A, and one end point is a new corner normal end point A with respect to the first after-moving outline line segment O8 and the second after-moving outline line segment O1. Perform expansion and contraction processing.

  FIG. 58 is a plan view showing the unit graphic shown in FIG. The first post-moving contour line segment O8 is extended so that the upper end thereof becomes the new corner normal end point A, and the second moving post contour line segment O1 is such that the left end thereof is the new corner normal end point A It is stretched to become. Thus, the two moving end points A1 and A2 shown in FIG. 57 are merged into the new corner normal end point A. If the same extension processing is performed on the remaining corner normal end points B, C, D, E, and G, a unit figure (corrected figure pattern 15) after correction as shown by a solid line in FIG. 58 is obtained.

  As illustrated, the unit figure after correction (corrected figure pattern 15 indicated by solid line) is a figure that is somewhat larger than the unit figure before correction (original figure pattern 10 indicated by broken line). Here, if the result of the simulation based on the original figure pattern 10 is correct and an appropriate correction is performed based on the result of the simulation, the actual lithography process is performed using this corrected figure pattern 15, the actual substrate S On the top, the same real figure pattern 20 as the original figure pattern 10 shown by a broken line is formed.

  After this correction figure pattern 15 (solid line), the process of moving the outline line segment described in 輪 郭 7.1 and the process of adding the outline line line described in § 7.2 with respect to the original figure pattern 10 (broken line) In this case, the pattern is constituted by a unit figure whose outline is a collection of outline line segments after the expansion and contraction process of the outline line segments described in § 7.3. However, when data of the data format shown in FIG. 55 is used to represent the unit graphic forming the original graphic pattern 10, the unit graphic forming the correction graphic pattern 15 is also data of the data format shown in FIG. 55. It will be expressed. The difference between the data indicating the original figure pattern 10 and the data indicating the corrected figure pattern 15 is only the position information (coordinate values) of the end point and the evaluation point included in each data pair shown in FIG.

<7.4 Repeated correction process>
In Section 7.3, an example in which the corrected figure pattern 15 indicated by a solid line in FIG. 58 is obtained by performing correction by the pattern correction unit 140 based on the original figure pattern 10 indicated by a broken line in FIG. Then, if an appropriate correction is made based on a correct simulation result, an actual lithography process is performed using the corrected figure pattern 15 to obtain the same actual figure pattern 20 as the original figure pattern 10 on the actual substrate S. Explained that they would be formed.

  However, in practice, it is always difficult to make one appropriate correction. For example, in “Contour line segment movement processing” of 7.1 7.1, when the estimated value of the process bias of a certain evaluation point is −1 nm, the correction process of moving the relevant evaluation point by +1 nm is exemplified. Such correction processing is not necessarily appropriate correction processing, and in fact, there may be cases where correction processing for moving the evaluation point by +1.1 nm is optimum correction processing.

  Although FIG. 1 shows an example in which a corrected figure pattern 15 can be obtained by giving the original figure pattern 10 to the shape correction apparatus 100, a lithography process is usually performed using the corrected figure pattern 15 thus obtained. Even when the actual figure pattern is formed on the actual substrate S, the actual figure pattern 20 obtained does not exactly match the original figure pattern 10 at the initial design stage. This is because the figures included in the original figure pattern 10 and the figures included in the corrected figure pattern 15 are different in size and shape, and therefore differences also occur in the proximity effect etc. when the lithography process is performed. .

  For this reason, in practice, as shown in FIG. 1, it is necessary to repeat the process of giving the corrected figure pattern 15 output from the pattern correction unit 140 to the figure correction device 100 again. That is, the corrected figure pattern 15 is given to the shape correction apparatus 100 instead of the original figure pattern, and the correction figure pattern 15 is subjected to the correction process again to obtain a new corrected figure pattern, and this new corrected figure pattern is obtained. Are again applied to the shape correction apparatus 100 to obtain a new corrected figure pattern.

  The repetition of such correction processing is also shown in the flow chart of FIG. That is, although the correction processing in steps S2 to S5 is initially performed on the original figure pattern created in step S1, the second and subsequent times are performed on the correction figure pattern obtained in step S5. Thus, the correction process of steps S2 to S5 is repeatedly executed until it is determined that the correction is completed in step S6. The actual steps of exposure, development and etching that constitute the lithography process of step S7 are performed based on the last corrected figure pattern finally obtained by such repeated processing.

  In order to repeat such correction processing, the feature quantity extraction unit 120 in the figure pattern shape correction device 100 shown in FIG. A function of extracting a feature indicating the feature around the evaluation point on the pattern 15 is provided, and the position on the corrected figure pattern 15 of the evaluation point on the actual figure pattern 20 is added to the bias estimation unit 130 based on the feature. The function of estimating the process bias that indicates the amount of deviation from the position of is performed, and the pattern correction unit 140 is further corrected with respect to the corrected figure pattern 15 based on the estimated value of the process bias output from the bias estimation unit 130 By providing a function to create a new corrected figure pattern There.

  Then, the correction on the corrected figure pattern 15 can be repeatedly performed, and a more appropriate corrected figure pattern can be created. When the correction is again performed on the corrected graphic pattern 15, the definition of the end point and the evaluation point is already completed, so the process by the evaluation point setting unit 110 can be omitted. That is, when the corrected figure pattern 15 drawn at the lower end of the block diagram shown in FIG. 1 is returned to the shape correction apparatus 100 again according to the upward arrow, it is not the evaluation point setting unit 110 but the feature quantity extraction unit 120 You can return it. The feature quantity extraction unit 120 performs feature quantity extraction processing for the existing evaluation points on the corrected figure pattern 15 returned, and the bias estimation unit 130 processes the process bias for the existing evaluation points on the corrected figure pattern 15 returned. What is necessary is just to obtain the estimated value y.

  In addition, when performing such correction processing, the contour line segment expansion and contraction processing unit 143 in the pattern correction unit 140 performs processing to expand and contract the contour line segments, and then the outline segment after the expansion and contraction processing It is preferable to have a function to perform processing for correcting the position of the evaluation point of.

  For example, in the extension and contraction processing of the outline segment described in § 7.3, the corrected figure pattern 15 shown in FIG. 58 is created. When the corrected figure pattern 15 shown in FIG. 58 is returned to the shape correction apparatus 100 again, the processing by the evaluation point setting unit 110 can be omitted as described above. This is because the end points A to H, F 'and H' and the evaluation points 1 to 8 have already been defined in the corrected figure pattern 15 shown in FIG. However, as illustrated, the positions of the evaluation points 1 to 8 on the corrected graphic pattern 15 are not necessarily appropriate positions. As described in 6.4 6.4, in order to obtain more appropriate simulation results, it is preferable to define each evaluation point at the midpoint of the contour line segment. However, when the expansion / contraction processing by the contour line segment expansion / contraction unit 143 is performed, the original evaluation point position deviates from the midpoint position of the contour line segment.

  Therefore, when the correction process is repeated, it is preferable that the contour line segment expansion and contraction processing unit 143 perform a process of correcting the position of the evaluation point after performing the process of stretching the contour line segment. Specifically, the position of each evaluation point may be corrected to the midpoint position of a new contour line segment. In FIG. 59, the contour line segment expansion and contraction processing unit 143 shown in FIG. 53 corrects the positions of the evaluation points 1 to 8 for each of the contour line segments O1 to O8 of the unit graphic after expansion and contraction processing shown in FIG. FIG. 6 is a plan view showing a broken state. As a result of this correction, the positions of the evaluation points 1 to 8 are at the midpoints of the contour line segments O1 to O8, so that more appropriate simulation can be performed thereafter.

<<< 8 8. Features and benefits of the present invention >>>
Finally, the features and advantages of the present invention will be described.

<Features of shape correction method according to the present invention>
Up to this point, the present invention has been described as an apparatus invention of a figure pattern shape correction apparatus, but the present invention can also be understood as a method invention of a figure pattern shape correction method. Thus, when the present invention is grasped as a method invention, when forming a real figure pattern on a real substrate by a lithography process based on the original figure pattern, the present figure makes the original figure pattern match the original figure pattern. This is a method of correcting the shape of a figure pattern in which the shape of the pattern is corrected to create a corrected figure pattern actually used in the lithography process.

  In this shape correction method, the evaluation point setting step (the evaluation point setting step mentioned above in which the computer sets an evaluation point at a predetermined position on an outline showing the boundary between the inside and the outside) for each unit figure constituting the original figure pattern A step performed by the unit 110), and a feature amount extraction step (a step executed by the feature amount extraction unit 120 described above) in which the computer extracts a feature amount indicating a feature around the evaluation point for the original figure pattern; Bias estimation step (performed by the bias estimation unit 130 described above, in which the computer estimates a process bias indicating the deviation between the position on the original figure pattern of the evaluation point and the position on the actual figure pattern based on the feature amount. Step) and a computer based on the estimated value of process bias obtained by this bias estimation step. There, by performing correction to the original figure pattern, and a pattern correction step of generating a correction figure pattern (steps performed by the above-mentioned pattern correction unit 140), the.

  Here, in the evaluation point setting step, a unit figure input step (step executed by the unit figure input unit 111 described above) for inputting individual unit figures composed of polygons, and division of the unit figures by straight dividing lines And a unit figure dividing step (step executed by the unit figure dividing unit 112 described above) for forming divided figures composed of a plurality of polygons, and for each divided figure, an outline of a polygon forming the divided figure. A divided figure vector defining step (step executed by the divided figure vector definition unit 113 described above) for defining divided figure vectors respectively on individual sides of the polygon so as to make a round along a predetermined forward direction along; From the set of divided figure vectors, overlapping vectors which are arranged on overlapping sides of a pair of adjacent polygons adjacent to each other and are oriented in the opposite direction to each other The unit graphic vector definition which defines a plurality of unit graphic vectors arranged along the contour of the unit graphic by searching the toll pair, removing the overlapping interval part of the searched overlapping vector pair, and remaining vectors Step (step executed by unit graphic vector definition unit 114 described above) and normal end point definition step defining normal end point at start position of each unit graphic vector respectively (executed by normal end point definition unit 115 described above And an evaluation point defining step (step executed by the evaluation point defining unit 117 described above) for defining an evaluation point between two ordinary end points arranged adjacent to each other on the contour line of the unit graphic. Search for an intermediate normal end point, which is a normal end point that does not constitute a polygon vertex, for each unit figure. And it has a middle normal replenishment endpoint definition step of defining additional supplemental endpoint in the same position as the end point (step executed by replenishing endpoint definition part 118 described above), the.

  In addition, in practice, in addition to the above-described steps, in the evaluation point setting stage, an end point reconstruction step in which the normal end point defined in the normal end point definition step is restructured to correct the number or position It is preferable that the steps executed by the end point reconstruction unit 116 be performed, and in the evaluation point definition step, the evaluation points be defined based on the normal end points obtained by the above reconstruction.

  On the other hand, in the pattern correction step, for each evaluation point, a normal end point or a replenishment end point adjacent to one side of the evaluation point along the contour of the unit figure is a first end point, and a normal end adjacent to the other side. Define a contour line segment whose end point or replenishment end point is the second end point, and the contour line segment for each evaluation point is orthogonal to the contour line segment by a correction amount according to the estimated value of the process bias of the evaluation point The contour line segment moving step (the step executed by the contour line segment movement processing unit 141 described above) which performs moving processing to move in the direction to move, and the intermediate normal end point and the replenishment end point that were at the same position before the moving processing, An outline line segment adding step (an outline line segment described above performs an addition process of adding a new outline line segment connecting the respective positions after movement process when separated into two different positions after movement process Step performed by the processing unit 142 and a corner normal end point, which is a normal end point that has constituted a vertex of the polygon before the moving process, is separated into two different positions after the moving process. An outline line segment expansion and contraction step for expanding and contracting each outline line segment including the corner normal end point after separation so that the two corner normal ends merge into a new corner normal end point And the steps executed by the contour line segment expansion and contraction processing unit 143.

  Then, in the shape correction method of a figure pattern according to the present invention, a corrected figure pattern configured by a unit figure having an outline of an aggregate of outline line segments after the movement process, the addition process, and the expansion process is performed. Is created. By adopting such a shape correction method, when performing a simulation necessary to correct the original figure pattern, efficient evaluation point setting can be performed that can reduce the calculation load as much as possible while maintaining simulation accuracy sufficient for practical use. It becomes possible.

<8.2 Advantages of Endpoint Definition Using Divided Figures>
Here, the merit in the exposure process obtained by adopting the present invention will be described. An important feature of the present invention is that the unit figure constituting the original figure pattern is divided into divided figures consisting of polygons, and the end points are defined with reference to the vertex positions of this polygon, and arranged adjacently. An evaluation point is defined for each contour line segment connecting two end points, and the position correction is performed for each contour line segment. By defining the evaluation points in this way, it is possible to define an appropriate number of evaluation points at appropriate positions. Therefore, when correction is performed on the original figure pattern, simulation accuracy sufficient for practical use is maintained. In addition, it is possible to perform efficient evaluation point setting that can reduce the calculation load as much as possible.

  In addition, defining the end point based on the vertex position of the polygon forming the divided figure provides an advantage that the total number of shots can be reduced in the exposure process. That is, the conventional general method adopts the method of defining the end points at predetermined intervals for each side of the unit figure, but such a conventional method has a disadvantage that the total number of shots increases. . In the present invention, such disadvantages can be eliminated. Hereinafter, this point will be described by giving a specific example.

  FIG. 60 is a plan view showing a procedure of a conventional method of defining end points at predetermined intervals for each side of a unit graphic. Now, when an original figure pattern including a unit figure 10 as shown in FIG. 60 (a) is given, a plurality of end points are defined on the outline of this unit figure 10, and an outline connecting the two adjacent end points Consider defining an evaluation point every minute and making corrections. In this case, first, the vertexes of the polygon constituting the unit graphic 10 are defined as the end points, and further, the end points are defined at predetermined intervals on each side of the polygon. Here, for convenience of explanation, the dimension of the vertical width of the unit graphic 10 is Wsplit, and on each side of the unit graphic 10, end points are defined at predetermined intervals of Wsplit or less.

  FIG. 60 (b) is an example in which the end points A to N are defined on the outline on the unit graphic 10 according to such a policy. Here, the end points A, E, F, G, H, L, M and N are the vertices of the polygon constituting the unit figure 10, while the remaining end points B, C, D, I, J and K are , Are the end points defined on the sides of the polygon. As described above, since the end points are defined such that the distance between adjacent end points is equal to or less than Wsplit, any end point distance is equal to or less than Wsplit, and an outline sandwiched between two adjacent end points The total length of the minutes is less than or equal to Wsplit.

  However, since the end points are defined independently for each side of the unit graphic 10, the end points B, C and D defined on the upper side of the polygon and the end point I defined on the lower side of the polygon are defined. There is no positional consistency at all between J and K. In other words, the lateral positions of the end points B, C, D and the lateral positions of the end points I, J, K are not aligned (see the broken line in FIG. 60 (b)).

  Since the shape correction for the unit graphic 10 is performed by moving the individual contour line segments, the unit graphic 15 after correction has, for example, a shape as shown in FIG. 60 (c). That is, the unit graphic 15 after correction shown in FIG. 60 (c) moves the contour line segments BC, CD, DE in the unit graphic 10 before correction shown in FIG. Is moved slightly outward, and the contour line segment JK is slightly moved inwardly.

  In a general writing apparatus used for a lithography process, an operation of exposing a trapezoidal area in one shot is adopted. In such a writing apparatus, a unit figure 15 after correction shown in FIG. 60 (c) is employed. In order to carry out the exposure process based on the above, it is necessary to divide the image into a number of rectangular areas as shown by the broken lines in the figure, and to perform exposure with each individual shot. That is, in the actual exposure process, it is necessary to perform individual shots for each of nine independent rectangles shown by hatching in FIG. 60 (d), and the total number of shots is nine. As described above, in the conventional method, since the positions of the end points defined on the opposite sides of the polygon forming the outline are not aligned, there is a disadvantage that the total number of shots increases when the outline is a regular figure ( If the contour is an anti-graphic, such disadvantages do not occur much).

  On the other hand, FIG. 61 is a plan view showing a procedure in the case of defining an end point based on the vertex position of a polygon constituting a divided figure by the method according to the present invention. The unit graphic 10 shown in FIG. 61 (a) is exactly the same as the unit graphic 10 shown in FIG. 60 (a), and here, according to the method of the present invention, a plurality of end points on the contour of the unit graphic 10 are Consider defining and correcting an evaluation point for each contour line segment connecting two adjacent end points.

  In this case, first, the unit figure 10 is divided into a plurality of divided figures. FIG. 61 (b) shows the unit figure 10 divided into five divided figures. As described above, the division process in the present invention is performed such that the size of each divided figure is equal to or less than the predetermined division interval Wsplit, so the dimensions of the vertical width and the horizontal width of the five divided figures shown in FIG. Both are below Wsplit. Then, an end point is usually defined at each vertex position of these five divided figures. FIG. 61 (b) shows a state in which the normal end points A to N are defined on the contour line on the unit graphic 10 according to such a policy. Here, the end points A, E, F, G, H, L, M, N are corner normal end points which are the vertices of the polygon constituting the unit graphic 10, and the remaining end points B, C, D, I , J, K are intermediate normal endpoints.

  Comparing FIG. 60 (b) with FIG. 61 (b), although all 14 end points A to N are defined in the same way, in the former, the end points B, C, and C defined on the upper side are the same. The lateral positions of the end points I, J and K defined on the lower side and the position D are not aligned and there is no consistency, whereas in the latter, the lateral positions of the upper and lower endpoints are aligned and the consistency is ensured. It is done.

  FIG. 61 (c) shows a unit figure 15 after correction obtained by correcting the unit figure 10 shown in FIG. 61 (b). In the unit figure 15 after this correction, the outline line segments BC, CD, DE in the unit figure 10 before the correction shown in FIG. 61 (b) are moved slightly inward, and the outline line segment IJ is slightly moved outward. The line segment JK is slightly moved inward. FIG. 61 (d) is a diagram showing individual rectangular areas divided to perform an exposure process based on the unit figure 15 after the correction.

  When dividing processing shown in FIG. 61 (b) is performed, dividing intervals Wsplit are set equal to or smaller than the beam maximum molding size Bmax of the drawing apparatus (setting of Wsplit ≦ Bmax), individual dividing shown in FIG. 61 (d). The size of the rectangular area is likely to be equal to or less than the beam maximum molding size Bmax. The exposure process is performed by performing individual shots for each of the individual rectangular areas, and the rectangular area having a size equal to or smaller than Bmax can be exposed with one shot by the drawing apparatus. In the case of the example shown in FIG. 61 (d), since any rectangular area shown by hatching has a size equal to or smaller than Bmax, in the actual exposure process, each of the five independent rectangular areas is 1 individually It is enough to do a shot.

  After all, in the case of the above-described embodiment, the conventional method requires a total of nine shots (see FIG. 60D) in the exposure step, whereas the method of the present invention requires a total of five shots. It will be finished. As described above, according to the present invention, since end points are defined using divided figures, there is obtained an advantage that efficiency of the exposure process (reduction of the total number of shots) can be achieved. Of course, according to the movement direction and movement amount of each contour line segment when creating the unit figure 15 after correction shown in FIG. 61 (c), the size of each rectangular area obtained after correction is the maximum beam forming size Bmax. In this case, the rectangular area can not be exposed in one shot. However, if the method according to the present invention is adopted, the total shot for the entire figure pattern including a large number of unit figures is obtained. The number will be statistically reduced.

<The merit of defining the replenishment end point>
In the present invention, as described above, an end point is usually defined using a division figure, and a replenishment end point is additionally defined at the same position for an intermediate normal end point which is a normal end point not constituting a polygon vertex. Here, the merit of defining the replenishment end point for the intermediate normal end point will be described.

  The reason for defining the replenishment end point in the present invention is to move the two sets of contour line segments located on both sides of one intermediate normal end point independently and independently as described in §6.5. For example, in the embodiment shown in FIG. 56, by additionally defining the replenishment end point F 'at the position of the intermediate normal end point F, the contour line segment O5 and the contour line segment O6 can be moved independently and independently. By additionally defining the replenishment end point H 'at the position of the intermediate normal end point H, it becomes possible to move the outline segment O7 and the outline segment O8 independently and independently.

  Moreover, in the present invention, the movement processing by the contour line segment movement processing unit 141 moves each evaluation point in a direction (normal direction: thick arrow direction in FIG. 56) orthogonal to the outline where the evaluation point is located. Based on the premise, the moving direction of the contour line segment to be moved parallel with the evaluation point is also called the normal direction. When such movement processing is performed, both the intermediate normal end point after movement and its replenishment end point will be on the normal to the position before movement. For example, in the case of the example shown in FIG. 56, the lateral positions of the end points F and F 'after movement do not change from before the movement, and the vertical positions of the end points H and H' after movement do not change from before the movement. Such features contribute to the efficiency of the exposure process (reduction of the total number of shots).

  FIG. 62 is a plan view showing the merit of defining the replenishment end point. Now, it is assumed that a rectangular unit graphic 10 as shown in FIG. 62 (a) is given, and shape correction is performed on this rectangular unit graphic 10 by the method according to the present invention. Here, it is assumed that this unit graphic 10 is a rectangle having a dimension of vertical width Wsplit and horizontal width 4 × Wsplit. Here, Wsplit is a division interval of the division processing performed by the unit graphic division unit 112, and for convenience of explanation, let us consider the case where Wsplit = Bmax, where the beam maximum molding size of the drawing apparatus is Bmax.

  FIG. 62 (b) is a plan view showing a state in which the unit graphic 10 shown in FIG. 62 (a) is divided by the unit graphic dividing section 112. As shown in the figure, the unit figure 10 is divided into four square-shaped divided figures shown by hatching, respectively. Usually, end points A to J (x marks) are defined. Further, the replenishment end point definition unit 118 sets the replenishment end points B ', C', D ', G', H ', I' (square marks) at the same positions with respect to the intermediate normal end points B, C, D, G, H, I. Is additionally defined.

  Here, as a result of performing correction processing by pattern correction unit 140 on unit figure 10 shown in FIG. 62 (b), unit figure 15 after correction as shown in FIG. 62 (c) is obtained. Let's do things. Specifically, the outline line segment movement processing unit 141 moves the outline line segments BC ', CD', GH ', HI' slightly inward, and the end point C 'is between the end points B'-B. The contour line segment addition processing unit 142 newly adds a section between −C, between the end points D ′ and D, between the end points G ′ and G, between the end points H ′ and H, and between the end points I ′ and I respectively. Contour line segments are added.

  However, since each contour line segment is moved in the normal direction, the end points B ', B, C', C, D ', D, G', G, H ', H, I' after movement processing. , I in the lateral direction is the same as the position before the movement process. Therefore, when performing the actual exposure process based on the unit figure 15 after this correction, if Wsplit = Bmax is set as described above, a bold square (one side is a drawing device in FIG. 62D) The exposure is completed by performing a total of four shots, as indicated by the square having a maximum beam forming size Bmax. Of course, in the correction process shown in FIG. 62 (c), when the movement process for moving each contour line segment to the outside of the figure is performed, a portion where the vertical width exceeds Bmax is generated, and it is not enough for a total of four shots. However, the total number of shots for the whole figure pattern including a large number of unit figures is statistically reduced.

  Such merits are also effective when repeating the correction processing described in 7.4 7.4. For example, after unit figure 15 after correction shown in FIG. 62 (c) is obtained by the first correction for unit figure 10 shown in FIG. 62 (a), unit figure 15 after this correction is further obtained. Consider the case where the unit figure 16 after correction shown in FIG. 63 (a) is obtained by performing the second correction. The unit graphic 16 shown in FIG. 63 (a) is obtained by moving the contour line segments BC ', CD', GH ', HI' of the unit graphic 15 shown in FIG. The lateral positions of the rear end points B ', B, C', C, D ', D, G', G, H ', H, I' and I are also the same as the positions before the movement processing.

  As shown in the example of the data format in FIG. 55, for the normal endpoints A, B, C, D, E, F, G and H, the corresponding evaluation points 1 to 8 are recorded, while the replenishment endpoint F is recorded. As for 'and H', empty data of evaluation points such as “Φ” which does not exist are recorded. This means that no evaluation point is defined on the contour line segment added by the contour line segment addition processing unit 142. Accordingly, in the case of the example shown in FIG. 63 (a), the contour line segments B'B, C'C, D'D, G'G, H'H, I'I added by the contour line segment addition processing unit 142. The evaluation points do not exist above, and no movement processing based on the movement of the evaluation point positions is performed on these contour line segments. In other words, these contour line segments are such that their lateral positions remain unchanged no matter how many correction processes are repeated.

  In the present invention, the meaning of defining the replenishment end point at the same position as the intermediate normal end point is that, when correction processing is performed, the positions of the intermediate normal end point and the replenishment end point change only in the normal direction of the outline. It is possible to realize an operation that does not change in the direction along the direction, and such an operation provides an advantage of making the exposure process more efficient (reducing the number of total shots).

  FIG. 64 is a plan view showing a disadvantage when the intermediate normal end point and the replenishment end point are moved in an arbitrary direction. Here, FIG. 64 (a) is a unit after correction obtained by performing correction to move the positions of the intermediate normal end point and the replenishment end point in an arbitrary direction with respect to the unit graphic 10 shown in FIG. 62 (b). A figure 17 is shown. In the unit graphic 15 after correction shown in FIG. 62 (c), the lateral positions of the end points B ', B, C', C, D ', D, G', G, H ', H, I', I are In the unit figure 17 after correction shown in FIG. 64 (a), the lateral position of these end points is changed, while it is unchanged.

  Therefore, when the actual exposure process is performed based on the unit figure 17 after correction shown in FIG. 64 (a), a bold line rectangle (one shot of the drawing apparatus having a maximum beam forming size Bmax in FIG. 64 (b) As shown by the rectangles of the size that can be exposed, the total of seven shots are required. As described above, according to the present invention, the replenishment end point is defined, and the movement direction of the intermediate normal end point and the replenishment end point is restricted to the normal direction of the contour line. It will be obtained.

<8.4 Advantages of Endpoint Reconstruction>
In the embodiments of the invention described above, reconstruction of the endpoints is performed as described in § 6.3. The reconstruction of this endpoint is not an essential requirement for practicing the present invention, but in practice, the reconstruction of the endpoint will improve the number and the position of the endpoint, so the evaluation will be more efficient. You can set points. Hereafter, the merit by this endpoint reconstruction is demonstrated based on a specific example.

  The L-shaped unit graphic 10 shown in FIG. 54 is subjected to end point definition and evaluation point definition as shown in FIG. 49 (b), and is shown by triangle marks by the end point reconstruction shown in FIG. Processing for removing the end point D is performed. When the correction process is performed on the unit figure 10 before the correction shown in FIG. 54, as described above, the unit figure 15 after the correction shown in FIG. 58 is obtained. The unit graphic 15 after correction has a shape obtained by enlarging the unit graphic 10 before correction to a certain extent, and is a graphic leaving the original L-shaped surface shadow.

  On the other hand, the unit figure 18 after correction shown in FIG. 65 is a figure obtained when the end point reconstruction shown in FIG. 43 is not performed. That is, the end point Z (triangular mark) shown in FIG. 65 corresponds to the end point D shown in FIG. 43, and the unit figure 18 after correction is obtained when it is assumed that the end point Z is not removed. It will be the figure after correction. Since the end point Z is an intermediate normal end point, if it remains without being removed by the end point reconstruction, the process of defining the replenishment end point will additionally define the replenishment end point Z '. The illustrated example is an example in which the intermediate normal end point Z has moved to the upper side of the figure and the replenishment end point Z 'has moved to the lower side of the figure. A contour line segment Z′Z is newly added by the process of adding the contour line segment performed after the movement process.

  Here, when the unit graphic 15 after correction shown in FIG. 58 and the unit graphic 18 after correction shown in FIG. 65 are compared, it can be seen that the contour in the vicinity of the vertex C is significantly changed in the latter. In general, in the case where the corner rounding effect (effect of rounding the corner portion near the apex of the polygon) is large when the lithography process is performed, the evaluation point is near the apex, the evaluation point The contour line segment including the & cir & is greatly fluctuated, and the contour line near the vertex is considerably fluctuated as in the example shown in FIG. If the outline near such a vertex changes significantly, it results in distorting the basic shape of the figure before correction, which is not preferable in practical use.

  The reconstruction of the end points described in 6.3 6.3 makes it possible to remove the end points close to the vertex of the unit shape, and as a result, it is possible to move the evaluation point away from the vertices. For this reason, it can suppress that the outline in the vicinity of a vertex changes remarkably, and the merit that it becomes possible to perform more efficient evaluation point setting is acquired.

Lastly, the table of contents for the section "Forms for carrying out the invention" is posted. The number [XXXX] after each heading is a paragraph number.
1. 1. Basic Configuration of Shape Correcting Device According to Prior Invention [0034]
1.1 Shape estimation device for figure pattern [0035]
1.2 Shape Correction Device for Graphic Pattern [0056]
1.3 Basic concept of feature quantity extraction in prior application invention [0071]
2. 2. Details of Feature Extraction Unit [0086]
2.1 Processing Procedure of Original Image Generating Unit 121 [0087]
2.2 Processing Procedure by Image Pyramid Creation Unit 122 [0103]
2.3 Processing Procedure of Feature Amount Calculation Unit 123 [0128]
2.4 Modification of Feature Extraction Processing [0144]
3. 3. Bias estimation unit details [0167]
3.1 Estimation operation by neural network [0168]
3.2 Learning stage of neural network [0183]
§4. Method for estimating shape of graphic pattern according to prior invention [0199]
5.5. Basic configuration of shape correction device according to the present invention [0205]
6.6. Details of Evaluation Point Setting Unit [0213]
6.1 Division of unit figure [0217]
6.2 Definition of vector and ordinary endpoints [0243]
6.3 Reconstruction of endpoint [0268]
6.4 Definition of Evaluation Points [0316]
6.5 Definition of replenishment end point [0321]
7.7. Details of pattern correction unit [0336]
7.1 Contour Line Segment Movement Processing [0339]
7.2 Contour line segment addition processing [0353]
7.3 Contour Line Segment Expansion Processing [0360]
7.4 Iteration of correction processing [0367]
§ 8. Features and merits of the present invention [0377]
8.1 Features of shape correction method according to the present invention [0378]
8.2 Advantages of Endpoint Definition Using Divided Figures [0384]
8.3 Advantages of Defining Refilling Endpoints [0397]
8.4 Advantages of Endpoint Reconstruction [0409]

  The shape correction apparatus and shape correction method of a figure pattern according to the present invention is a lithography process based on an original figure pattern in a field where fine patterning needs to be performed on a specific material layer, such as a semiconductor device manufacturing process. When forming a real figure pattern on a real substrate, it can be widely used as a technique for forming a real figure pattern having accurate dimensions.

1 to 18: Evaluation point 10: Original figure pattern (unit figure before correction)
10a to 10l: divided figure 10X: figure body 10Y: hole opening 15: corrected figure pattern (unit figure after correction)
15a to 10d: corrected divided figures 16, 17 and 18: corrected unit figure 20: actual figure pattern 100: figure pattern shape correction apparatus 100 ': figure pattern shape estimation apparatus 110: evaluation point setting unit 111: Unit graphic input unit 112: unit graphic division unit 113: divided graphic vector definition unit 114: unit graphic vector definition unit 115: normal end point definition unit 116: end point reconstruction unit 117: evaluation point definition unit 118: replenishment end point definition unit 120: Feature extraction unit 121: original image creation unit 122: image pyramid creation unit 123: feature amount calculation unit 130: bias estimation unit 131: feature amount input unit 132: estimation operation unit 140: pattern correction unit 141: contour line segment movement processing Unit 142: Contour line segment addition processing unit 143: Contour line segment expansion / contraction processing units A to D: individual pixels / individual pixels Pixel values A to Z: end points a to d: lateral distance or longitudinal distance A to A4 between evaluation point E and center point of each pixel: forward arrow A0: corner normal end points A1 and A2 before movement processing: Corner normal point Bmax after movement processing: Beam maximum molding size b, b (1, 1) to b (i + 1, M (i + 1), b (N + 1): parameters of neural network C1, C2: reference circles D1 to Dn: difference image E, E11 to E23: evaluation points F1 to F5: figures (rectangles) constituting the original figure pattern 10
f (ξ): Function G used for calculation of neural network G: proportional value GF 33, GF 55: Gaussian filter H: proportional value h (1, 1) to h (N, M (N)): neuron in hidden layer of neural network / Calculated value i: Parameter J indicating the number of hidden layers of neural network: intersection point k of two straight lines: parameter L indicating image number: learning information LF33, LF55: Laplacian filter L1, L2, L3: distance Lcd between end points , Lde, Lij: Distance between end points Leven: Distance between adjacent end points (equally spaced)
Lodd: distance between adjacent end points (fractional distance)
Lreg: distance between adjacent end points (reference distance)
M1: area density map M2: edge length density map M3: dose density map M (1) to M (N): dimension N of each hidden layer of neural network: number of stages of hidden layer of neural network / total number of end points n: image Number of layers of pyramid / number of end points O, O1 to O8, O5 ', O7': contour line segment Oa, Ob: contour line segment Oin: inner contour line (anti-figure)
Oout: Outer contour (positive figure)
P1 to Pn: hierarchical image Pk: kth hierarchical image PC: corrected image PD: image pyramid (sub image pyramid)
PP: Image pyramid (main image pyramid)
Q1: The first preparation image (original image)
Qk: kth preparation image S: real substrate S1 to S848: steps SS1 to SS44 of the flowchart: steps T1, T2, T (n-1), T (n), T (n + 1), T of the flowchart (N + 2): normal end points T ', T (n)', B 'to Z', replenishment end point U: pixel u: pixel dimensions V1 to V20: unit figure vectors Va1 to Vh4: divided figure vectors W, W (1 , 11) to W (N + 1, 1 M (N)): parameter Wmerge of neural network: fusion reference value Wrest: width of remaining portion Wsplit: division interval X: coordinate axis x on two-dimensional XY plane x, x1 to xn: feature amount Y: coordinate axes y, y11 to y13 on the two-dimensional XY plane: process bias Z: deleted end point ξ: arguments of function f ξ 1 to ξ 4: vertex of divided figure :: empty data of evaluation point not existing

Claims (21)

When the real figure pattern (20) is formed on the real substrate (S) by the lithography process based on the original figure pattern (10), the real figure pattern (20) matches the original figure pattern (10) A figure pattern shape correction apparatus (100) for correcting the shape of the original figure pattern (10) to create a corrected figure pattern (15) actually used in the lithography process,
An evaluation point setting unit (110) for setting an evaluation point (E) at a predetermined position on an outline indicating the boundary between the inside and the outside for each unit graphic (10) constituting the original graphic pattern (10);
A feature quantity extraction unit (120) for extracting feature quantities (x1 to xn) indicating features around the evaluation point (E) for the original graphic pattern (10);
Process bias (y) indicating the amount of deviation between the position of the evaluation point (E) on the original figure pattern (10) and the position on the actual figure pattern (20) based on the feature quantities (x1 to xn) A bias estimation unit (130) for estimating
A pattern correction unit (15) for creating a corrected figure pattern (15) by performing correction on the original figure pattern (10) based on the estimated value (y) of the process bias output from the bias estimation unit (130) 140),
Equipped with
The evaluation point setting unit (110)
A unit figure input unit (111) for inputting individual unit figures (10) consisting of polygons;
A unit figure dividing unit (112) for dividing the unit figure (10) by a straight dividing line to form divided figures (10a to 10l) composed of a plurality of polygons;
For each of the divided figures (10a to 10l), divided figure vectors (Va1) are respectively placed on the respective sides of the polygon so as to make a round around a predetermined forward direction along the outlines of the polygons constituting the divided figures. A divided figure vector definition unit (113) that defines ~ Vh4),
The set of divided figure vectors (Va1 to Vh4) is arranged on overlapping sides of a pair of adjacent polygons adjacent to each other, and the overlapping vector pairs whose directions are opposite to each other are searched, and the searched overlapping vector pairs Unit graphic vector definition unit (114 for defining a plurality of unit graphic vectors (V1 to V20) arranged along the outline of the unit graphic (10) by removing the overlapping interval part of )When,
A normal end point definition unit (115) that defines a normal end point (T) at the start position of each unit graphic vector (V1 to V20);
An evaluation point definition unit (117) defining an evaluation point (E) between two ordinary end points (T) arranged adjacent to each other on the contour line of each unit graphic (10);
For each unit figure (10), search for an intermediate normal end point which is a normal end point (T) not constituting a polygon vertex, and additionally define a replenishment end point (T ') at the same position as the searched intermediate normal end point A replenishment end point definition unit (118),
Have
The pattern correction unit (140)
For each evaluation point (E), a normal end point (T) or a replenishment end point (T ') adjacent to one side of the evaluation point (E) along the contour of the unit graphic (10) is taken as the first end point Define a contour line segment whose second end point is the normal end point (T) or the replenishment end point (T ') adjacent to the other side, and the contour line segment for each evaluation point (E) is evaluated A contour line segment movement processing unit (141) that performs movement processing of moving in a direction orthogonal to the contour line segment by the correction amount according to the estimated value (y) of the process bias of the point (E);
If the intermediate normal end point and replenishment end point (T ') that were at the same position as each other before the movement processing are separated into two different positions after the movement processing, new contour line segments connecting the positions after each movement processing are A contour line segment addition processing unit (142) that performs additional processing to be added;
If the normal edge (T), which is the normal end point (T) that made up the vertex of the polygon before the movement process, separates into two different positions after the movement process, the two corner normal points after separation merge A contour line segment expansion and contraction processing unit (143) that performs expansion and contraction processing to expand and contract each contour line segment including the corner normal point after separation so as to be a new corner normal point;
Creating a corrected figure pattern (15) composed of a unit figure (10) whose outline is a collection of contour line segments after the movement process, the addition process, and the expansion / contraction process. A figure pattern shape correction apparatus (100) characterized by In the figure correction apparatus (100) of the figure pattern according to claim 1,
A graphic pattern shape correction apparatus (100), characterized in that a unit graphic division unit (112) divides a unit graphic (10) to form a trapezoidal divided graphic. The shape correction apparatus (100) for a graphic pattern according to claim 1 or 2,
When the unit graphic input unit (111) inputs an elongated strip unit graphic (10) extending along a predetermined longitudinal axis and having linear upper and lower sides parallel to each other,
A unit graphic division unit (112) divides the strip unit graphic unit (10) by a plurality of division lines arranged at predetermined intervals along the longitudinal axis, thereby one-dimensionally along the longitudinal axis. A figure pattern shape correction apparatus (100) characterized by forming a plurality of divided figures arranged in a row. The shape correction apparatus (100) for a graphic pattern according to claim 1 or 2,
When the unit graphic input unit (111) inputs a unit graphic (10) defined on the XY plane,
The unit graphic division unit (112) arranges a plurality of horizontal division lines parallel to the X axis at predetermined intervals in the Y axis direction, and divides the unit graphic (10) by the plurality of horizontal division lines. By forming a plurality of elongated strip shapes extending in the X-axis direction, and further dividing each of the plurality of strip shapes by a plurality of vertical dividing lines disposed at predetermined intervals along the X axis and parallel to the Y axis A graphic pattern shape correction apparatus (100), comprising: forming a plurality of divided figures two-dimensionally arranged. In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 4,
The unit figure dividing unit (112) has a function of setting the maximum beam forming size of the drawing apparatus used in the lithography process, and the arrangement interval of the dividing line used for dividing the unit figure (10) is less than the beam maximum forming size A graphic pattern shape correction apparatus (100) characterized by setting. In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 5,
The divided figure vector definition unit (113) assigns the numbers from the first vertex to the Kth vertex in the order of clockwise or counterclockwise along the outline to each vertex of the divided figure composed of K-gon, The process of defining the kth divided figure vector (Va1 to Vh4) having the vertex of k as a start point and the (k + 1) th vertex (where the first vertex in the case of k = K) as an end point is k = A figure pattern correction device (100) for defining a divided figure vector for the divided figure by repeating the process from 1 to K repeatedly. In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 6,
Graphic pattern shape correction apparatus characterized in that an evaluation point definition unit (117) defines an evaluation point at a midpoint between two normal end points arranged adjacent to each other on the contour line of the unit graphic (10). (100). In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 7,
The replenishment end point definition unit (118) defines a total of N normal end points defined for one outline, in the order along the outline, from the first normal end point T (1) to the Nth normal end point. When calling T (N), focusing on the n-th normal endpoint T (n), the (n-1) -th normal endpoint T (n-1) (where n = 1, the Nth N) End point T (N), the n-th normal end point T (n), and the (n + 1) -th normal end point T (n + 1) (where n = N, the first normal end point T (1)) If the nth normal end point T (n) is determined as an intermediate normal end point when three points are on a straight line, a process of additionally defining a replenishment end point at the same position as the intermediate normal end point is n = 1 to N A graphic pattern shape correction apparatus (100) characterized by repeatedly executing the process. In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 8,
When the outline segment extension / contraction processing unit (143) separates one corner normal end point into a first moving end point and a second moving end point by moving processing, the first moving end point is one end point Find a point of intersection between the straight line including the first post-moving contour line segment and the straight line including the second post-moving contour line segment having the second moving end point as one end point, and the intersection point is a new corner A normal end point is performed, and an expansion / contraction process is performed such that one end point becomes the new corner normal end point with respect to the first post-moving outline line segment and the second post-moving outline line segment. Shape correction apparatus (100) for graphic patterns. In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 9,
An end point restructuring unit (Evalence point setting unit (110) performs reconstruction to correct the number or the position of the normal end points defined on the outline of the unit graphic (10) by the normal end point definition unit (115) 116) further,
A figure pattern shape correction apparatus (100), wherein an evaluation point definition unit (117) defines an evaluation point based on the normal end point obtained by the reconstruction. In the shape correction apparatus (100) of the figure pattern according to claim 10,
The end point reconstruction unit (116) is characterized in that the end point (T) is normally reconstructed by an end point removal process for removing a part or all of the intermediate normal end points with respect to a part or all unit graphics (10). Shape correction device (100) of a figure pattern. In the figure correction apparatus (100) of the figure pattern according to claim 11,
The end point reconstruction unit (116) generates a total of N normal end points defined for one outline of one unit graphic (10) in the order along the outline, the first normal end point T. (1) to N-th normal end points T (N), focusing on the n-th normal end point T (n), the n-th normal end point T (n) and the removal target located immediately before it The normal end point T (n-m) (where m is a natural number, and in the case of n-m <1, the distance to T (n-m + N) is L1, and the n-th normal end point T (n Assuming that the distance between n) and the (n + 1) th normal end point T (n + 1) (where n = N, the first normal end point T (1): the same applies hereinafter) is L2
At least one of L1 and L2 is equal to or less than a predetermined reference value Wmerge, and the normal end point T (n-1) (where n = 1, the Nth normal end point T (N)), T (n) , T (n + 1), the determination process for the n-th normal end point T (n) is repeatedly executed until n = 1 to N, and becomes the removal target. A figure pattern shape correction device (100) characterized by performing end point removal processing for removing an end point normally. In the figure correction apparatus (100) of the figure pattern according to claim 11,
The end point reconstruction unit (116) generates a total of N normal end points defined for one outline of one unit graphic (10) in the order along the outline, the first normal end point T. (1) to N-th normal end points T (N), n-th normal end points T (n), and normal end points T (n-m) not immediately before that and not to be removed Is a natural number, and in the case of n−m <1, the distance to T (n−m + N) is L1 and the n th normal endpoint T (n) and the (n + 1) normal endpoint T (n + 1) (However, in the case of n = N, let L2 be the distance between the first normal end point T (1) and so on) and the (n + 1) th normal end point T (n + 1) and the (n + 2) th normal end point T (n + 2) (however, the first normal end point T (1) in the case of n = N-1; the second normal end point T (2) in the case of n = N: the same applies hereinafter) ) The distance between the when the L3,
L2 is less than or equal to a predetermined reference value Wmerge, and the normal end point T (n-1) (where n = 1, the Nth normal end point T (N): the same applies hereinafter), T (n), T When the four points (n + 1) and T (n + 2) lie on a straight line, the normal end point T (n) is targeted for removal if L1 <L3, and the normal end point T (n + 1) if L1> L3. If L1 = L3, then either normal endpoint T (n) or normal endpoint T (n + 1) is targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n-1), T (n) and T (n + 1) are on a straight line, but the normal end point T (n + 2) is If it is not on the line, the end point T (n) is usually targeted for removal,
L2 is less than or equal to a predetermined reference value Wmerge, and normally three end points T (n), T (n + 1) and T (n + 2) are on a straight line, but the normal end point T (n-1) is a straight line If not on the line, the end point T (n + 1) is usually targeted for removal,
The figure correction apparatus characterized in that an end point removing process is performed to repeat the process above while incrementing n by 1 until n N N is satisfied and removing the normal end point to be removed. 100). In the shape correction apparatus (100) of the figure pattern according to claim 10,
The end point reconstruction unit (116) is a normal end point by an end point rearrangement process in which an intermediate normal end point is once removed for a part of the outline and a new intermediate normal end point is placed on each side of the part of the outline. A shape correction apparatus (100) for a figure pattern, characterized by performing reconstruction of In the figure correction apparatus (100) of the figure pattern according to claim 14,
The end point repositioning process is performed by the end point reconstruction unit (116) removing intermediate normal end points for a specific outline, and then repositioning the intermediate normal end points at predetermined intervals on each side of the specific outline. A shape correction device (100) for a figure pattern characterized by performing. In the shape correction apparatus (100) of the figure pattern according to claim 10,
The end point reconstruction unit (116) performs the end point removal process according to any one of claims 11 to 13 and the end point relocation process according to claim 14 or 15 for each outline of the unit figure (10). Have the ability to run selectively,
End points are reconstructed by the end point removing process for normal end points on a regular figure surrounded by unit figure vectors (V1 to V20) around the forward direction, and a unit around the reverse direction opposite to the forward direction A figure pattern correction device (100) of figure pattern, wherein end point reconstruction is performed by the end point rearrangement process for normal end points on an anti-figure surrounded by figure vectors (V1 to V20). In the figure correction apparatus (100) of the figure pattern according to any one of claims 1 to 16,
For the corrected figure pattern (15) created by the pattern correction unit (140) by the feature quantity extraction unit (120), a feature quantity (E) indicating the feature around the evaluation point (E) on the corrected figure pattern (15) have a function of extracting x1 to xn),
The bias estimation unit (130) determines the deviation between the position of the evaluation point (E) on the corrected figure pattern (15) and the position on the actual figure pattern (20) based on the feature quantities (x1 to xn). Have the ability to estimate process bias (y)
A new correction is made by the pattern correction unit (140) further correcting the corrected figure pattern (15) based on the estimated value of the process bias (y) output from the bias estimation unit (130). Has the function of creating a graphic pattern (15),
A figure pattern shape correction device (100) characterized in that correction on a figure pattern (15) is repeatedly executed. In the figure pattern shape correction device (100) according to claim 17,
The contour line segment expansion / contraction processing unit (143) is characterized by performing processing to expand or contract the contour line segment, and then to execute processing to correct the position of the evaluation point (E) with respect to the contour line segment after expansion / contraction processing. Shape correction apparatus (100) for graphic patterns.   The computer as the shape correction device (100) for a graphic pattern according to any one of claims 1 to 18 or an evaluation point setting unit (110) or a pattern correction unit (140) included in the shape correction device (100) for the graphic pattern A program that makes When the real figure pattern (20) is formed on the real substrate (S) by the lithography process based on the original figure pattern (10), the real figure pattern (20) matches the original figure pattern (10) And correcting the shape of the original figure pattern (10) to form a corrected figure pattern (15) actually used in the lithography process.
An evaluation point setting step (S2) in which a computer sets an evaluation point (E) at a predetermined position on an outline indicating the boundary between the inside and the outside for each unit graphic (10) constituting the original graphic pattern (10) )When,
A feature amount extraction step (S3) in which a computer extracts feature amounts (x1 to xn) indicating features around the evaluation point (E) for the original figure pattern (10);
A process for showing a deviation between the position of the evaluation point (E) on the original figure pattern (10) and the position on the actual figure pattern (20) based on the feature quantities (x1 to xn) A bias estimation step (S4) of estimating a bias (y);
Pattern correction to create a corrected figure pattern (15) by performing correction on the original figure pattern (10) based on the estimated value (y) of the process bias obtained by the bias estimation step (S4). Stage (S5),
Have
The evaluation point setting step (S2) is
A unit figure input step (111) for inputting individual unit figures (10) consisting of polygons;
A unit figure dividing step (112) of dividing the unit figure (10) by a straight dividing line to form a divided figure consisting of a plurality of polygons;
For each of the divided figures, the divided figure vectors (Va1 to Vh4) are defined on the respective sides of the polygon so as to make a round in a predetermined forward direction along the outline of the polygon constituting the divided figure. Divided figure vector defining step (113)
The set of divided figure vectors (Va1 to Vh4) is arranged on overlapping sides of a pair of adjacent polygons adjacent to each other, and the overlapping vector pairs whose directions are opposite to each other are searched, and the searched overlapping vector pairs Step of defining unit figure vector defining the plurality of unit figure vectors (V1 to V20) arranged along the outline of the unit figure (10) by removing the overlapping section part of )When,
A normal end point defining step (115) of defining normal end points (T) at start positions of individual unit graphic vectors (V1 to V20);
An evaluation point defining step (117) for defining an evaluation point (E) between two ordinary end points (T) arranged adjacent to each other on the contour line of the unit graphic (10);
For each unit figure (10), a replenishment end point defining step (118) which searches for an intermediate normal end point which is a normal end point not constituting a polygon vertex, and additionally defines a replenishment end point at the same position as the searched intermediate normal end point. When,
Have
The pattern correction step (S5)
For each evaluation point (E), a normal end point (T) or a replenishment end point (T ') adjacent to one side of the evaluation point (E) along the contour of the unit graphic (10) is taken as the first end point Define a contour line segment whose second end point is the normal end point (T) or the replenishment end point (T ') adjacent to the other side, and the contour line segment for each evaluation point (E) is evaluated A contour line segment moving step (141) that performs a movement process of moving the point (E) in a direction orthogonal to the contour line segment by the correction amount according to the estimated value (y) of the process bias;
If the intermediate normal end point and replenishment end point (T ') that were at the same position as each other before the movement processing are separated into two different positions after the movement processing, new contour line segments connecting the positions after each movement processing are A contour line segment adding step (142) that performs additional processing to be added;
When the corner normal end point, which is a normal end point that usually constitutes the vertex of the polygon before the movement processing, is separated into two different positions after the movement processing, the two corner normal ends after separation merge to form a new A contour line segment expansion / contraction step (143) for performing expansion / contraction processing for respectively expanding and contracting each contour line segment including the corner normal end point after separation so as to be a corner normal end point;
Creating a corrected figure pattern (15) composed of a unit figure (10) whose outline is a collection of contour line segments after the movement process, the addition process, and the expansion / contraction process. A method of correcting the shape of a figure pattern characterized by In the shape correction method of a graphic pattern according to claim 20,
The evaluation point setting step (S2) further includes an end point reconstruction step (116) that performs reconstruction to correct the number or position of the normal end point (T) defined by the normal end point definition step (115). ,
In the evaluation point definition step (117), an evaluation point (E) is defined based on the normal end point (T) obtained by the reconstruction, and the figure pattern correction method.

Images