JPH05101166A - Pattern matching device - Google Patents

Abstract

PURPOSE:To easily perform pattern matching in a short time by finding a secondary electron image magnification error and a sample positioning error which maximize the degree of pattern matching when edge positions are varied according to the secondary image magnification precision, sample positioning precision, and wiring width expansion or reduction. CONSTITUTION:This pattern matching device is equipped with edge position detecting means 3, secondary electron image storage means 4, a projection brightness distribution generating means 5, edge degree detecting means 6, a pattern matching degree calculating means 7, and an error detecting means 8. An edge degree E(X) is found from the brightness distribution B(X) obtained by projecting a secondary electron image on an axis X and the degree V of pattern matching between the edge degree E(X) and linear edge positions X1 and i=1-(n) on a CAD data is found to find the secondary electron image magnification error and sample positioning error which maximize the degree of pattern matching when the edge positions X1 and i=1-(n) are varied according to the secondary electron image magnification precision, sample positioning precision, and wiring width expansion or reduction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pattern matching device for performing pattern matching between a wiring pattern of a secondary electron image obtained by scanning an electron beam on a sample with an electron beam device and a wiring pattern on CAD data. Regarding

[0002]

2. Description of the Related Art When measuring the voltage of wiring on a semiconductor chip with an electron beam tester, the surface of the chip is scanned with an electron beam to obtain a secondary electron image, which is displayed on a display device. After visually observing the image and designating a measurement point, the voltage is measured by irradiating the point with an electron beam.

However, it is not easy to specify the measurement point on the secondary electron image for the following reason.

With the increase in the scale and integration of semiconductor integrated circuits, the field of view of the secondary electron image becomes relatively narrow, and the positioning error of the stage on which the sample is mounted is relatively large with respect to the wiring pitch. It's getting bigger.

In the secondary electron image, the wiring pattern of the uppermost layer is clear, but the wiring pattern one layer below the uppermost layer has a low contrast and is not clear.

Therefore, there has been proposed a method of automatically converting a measurement point designated on the basis of CAD layout data of a wiring mask pattern into a measurement point on a secondary electron image by pattern matching between a layout diagram and a secondary electron image. There is.

[0007]

However, this pattern matching is not easy because of the following reasons.

There is a large amount of image data, for example, 512 × 512 pixels × 8 bits in only a secondary electron image.

The secondary electron image is brightness data for each pixel, whereas the CAD data is polygon data (vector data).

The accuracy of the actual magnification with respect to the set magnification of the secondary electron image is about ± 2%, and the positioning accuracy of the stage on which the sample is mounted is about ± 2 μm.

Depending on the conditions of the semiconductor chip manufacturing process, the wiring is thickened or thinned, and the degree of thickening or thinning is different for each wiring layer.

In normal pattern matching, the product sum of corresponding pixels is calculated as the pattern matching degree each time the image is shifted, and the shift amount that maximizes the pattern matching degree may be obtained. Each time the magnification and the degree of thickening or thinning of each wiring are changed, the image is shifted to calculate the pattern matching degree, and the magnification, the degree of thickening or thinning of the wiring, and the shift amount that maximize the pattern matching degree must be obtained. Therefore, the processing is complicated and the processing time becomes long.

In view of the above problems, an object of the present invention is to provide a pattern matching device which can easily and quickly perform pattern matching between a wiring pattern on CAD data and a wiring pattern on a secondary electron image. To provide.

[0014]

FIG. 1 is a diagram showing the principle of the pattern matching device according to the present invention. This pattern matching device performs pattern matching between a wiring pattern of a secondary electron image obtained by scanning an electron beam on a sample with an electron beam device and a wiring pattern on CAD data. It has various components.

Numeral 3 is an edge position detecting means for the CAD
The positions X _i , i = 1 to n of the wiring pattern edge parallel to one axis Y of the orthogonal coordinate system on the data are detected. A secondary electron image storage means 4 stores the secondary electron image. 5
Is a projection brightness distribution creating means, which obtains a projection brightness B (X) obtained by adding the brightness of the secondary electron image along a straight line parallel to one axis Y of the orthogonal coordinate system on the secondary electron image. In this case, the range of Y may be limited based on the secondary electron image magnification accuracy, sample positioning accuracy, and wiring width scaling. An edge degree detecting unit 6 detects the edge degree E (X) of the wiring pattern from the projection brightness B (X). Reference numeral 7 denotes a pattern matching degree calculation means, which obtains the degree of correlation between the edge positions X _i , i = 1 to n and the edge degree E (X) as the pattern matching degree V. Reference numeral 8 is an error detecting means,
Edge position X _i of wiring pattern on data, i = 1 to n
The secondary electron that maximizes the pattern matching degree V when is changed in the range of X _i −a _i ≦ X _i ≦ X _i + b _i based on the secondary electron image magnification accuracy, sample positioning accuracy, and wiring width scaling. Determine the image magnification error and sample positioning error. It is not necessary to consider all wiring patterns for the pattern matching degree V, and the detection range of the edge positions X _i , i = 1 to n may be appropriately limited. For example, the edge position may be limited to only the edge position of the uppermost wiring pattern.

In the present invention, the edge degree E (X) is obtained from the brightness distribution B (X) obtained by projecting the secondary electron image on the axis X, and the edge degree E (X) and the one-dimensional edge position on the CAD data are obtained. X _i , the pattern matching degree V with i = 1 to n is obtained,
The edge position X _i , i = 1 to n, the secondary electron image magnification accuracy,
Since the secondary electron image magnification error and the sample positioning error which maximize the pattern matching degree when changed based on the sample positioning accuracy and the wiring width scaling, the wiring pattern on the CAD data and the wiring pattern on the secondary electron image are obtained. It becomes possible to perform pattern matching with and easily, accurately and in a short time.

In the first aspect of the present invention, the pattern matching degree calculation means 7 sets the pattern matching degree V to V = E.
It is calculated by (X ₁ ) + E (X ₂ ) + ... + E (X _n ).

Let Σ _X be the sum of all X's in the specified area, and let e (X) be 1 if an edge exists at the position X on the CAD data, and 0 if it does not exist. V = sigma _X is E (X) e (X) = E (X 1) + E (X 2) + ··· + E (X n), also the edge position X _i, i = 1 to n is CAD data According to this configuration, the pattern matching degree V can be accurately and easily evaluated, since it can be easily detected.

In the second aspect of the present invention, the edge degree detecting means 6 determines the edge degree E (X) as E (X) = | B (X-q)-.
It is calculated by B (X) | + | B (X) -B (X + q) |.

With this configuration, the edge of the wiring pattern can be correctly evaluated regardless of the wiring layer, the right edge, and the left edge.

In the third aspect of the present invention, the projection luminance distribution creating means 5 uses the edge position X _i , i = 1 to n range X _i −a _i.
The projection brightness B (X) is obtained only for ≤X _i ≤X _i + b _i .

In the case of this configuration, the data to be processed becomes the minimum necessary, and the data can be processed in a shorter time.

In the fourth aspect of the present invention, the edge position detecting means 3 also detects the edge straight line of the oblique wiring pattern which is non-parallel to both axes of the orthogonal coordinate system on the CAD data, and the projection brightness distribution creating means 5 Also obtains the distribution of the oblique projection brightness obtained by adding the brightness of the secondary electron image along the straight line on the secondary electron image parallel to the edge straight line, and the edge degree detecting means 6 determines the oblique projection brightness distribution. The pattern matching degree calculating means 7 also obtains the degree of correlation between the edge straight line and the edge degree and adds it to the pattern matching degree V, and the error detecting means 8 The edge straight line of the diagonal wiring pattern on the CAD data is also changed based on the secondary electron image magnification accuracy, the sample positioning accuracy, and the wiring width expansion / contraction to obtain the secondary electron image magnification error and the maximum degree of pattern matching V. Trial Determine the positioning error.

By appropriately adopting the orthogonal coordinate system, most of the wiring patterns are parallel to one of both axes of the orthogonal coordinate system, and even if there is a diagonal wiring pattern which is not parallel to both axes, this is ignored. Although the pattern matching can be performed by using the fourth aspect, the pattern matching process can be performed on the diagonal wiring pattern in the same manner as other wiring patterns.

[0025]

Embodiments of the present invention will be described below with reference to the drawings.

(1) First Embodiment FIG. 2 shows that by designating a measurement point based on the design data of the wiring pattern, the corresponding measurement point on the secondary electron image is automatically determined and the voltage is measured. The hardware structure of an electron beam tester is shown.

The electron beam irradiation device 10 detects the amount of secondary electrons according to the voltage at the measurement point and the voltage applied to the energy analysis grid, and the electron gun 13 is applied to the sample 12 placed on the stage 11. When the electron beam EB is emitted from the electron beam EB, the electron beam EB is emitted from the condenser magnetic field lens 14
A, pulsed through the blanking deflector 15 and blanking aperture 16, pulsed through the condenser magnetic field lens 14B, the deflector 17 and the objective magnetic field lens 18, and convergently irradiated to the measurement point on the sample 12, and the irradiation point The secondary electrons SE emitted from the detector pass through the extraction grid 191, the control grid 192, and the energy analysis grid 193, and are detected by the secondary electron detector 20.

The electron beam irradiation point on the sample 12 is scanned by the drive voltage supplied from the deflection controller 21 to the deflector 17. The magnification of the secondary electron image is determined by the amplitude E0 of this drive voltage. The deflection control device 21 uses the electron beam E
The scanning counter 21a corresponding to the B scanning position is provided, and the clock count value A thereof is supplied to the image input device 22. For example, if A is 18 bits, and the upper 9 bits are y and the lower 9 bits are x, a voltage E ₀ x / 511 is applied to the X deflection plate of the deflector 17, and the Y deflection plate of the deflector 17 is applied. A voltage E ₀ y / 511 is applied to the sample 12 and
The coordinates (X, Y) of the electron beam irradiation point above are (E ₀ kx, E 0ky) in an ideal case. Here, k is a constant.

The image input device 22 includes the secondary electron detector 20.
After amplifying the output of, it is converted into a digital value and this is the brightness L
Is written in the address A of the SEM image frame memory 23. The contents of the SEM image frame memory 23 are supplied to the display device 24, the secondary electron image is displayed on the screen thereof, and read by the computer 25 to be image-processed as described later.

On the other hand, the CAD data storage device 27 stores CAD data for obtaining the photomask of the wiring pattern. As will be described later, the computer 25 sends the CA data from the CAD data storage device 27 based on the designated range.
The D data is read out, the set magnification M _S of the secondary electron image is determined, this is set in the deflection control device 21, the target position of the stage 11 is determined, and this is set in the stage control device 26, and the secondary electron is set. Pattern matching between the image and the CAD data is performed to determine the measurement point on the secondary electron image from the measurement point on the CAD data.

The measurement points and specified range on the layout diagram are
It is supplied from an input device 28 such as a keyboard and a storage device, and this input data and other data are displayed on the display device 29. Further, the secondary electron image setting magnification M _S is supplied to the deflection controller 21, and the deflection controller 21 applies the drive voltage having the amplitude E 0 inversely proportional to the magnification M _S to the deflector 17.
Supply to.

Next, referring to FIG. 3, the computer 25
An outline of the voltage measurement procedure by will be described. Hereinafter, the numerical value in the parenthesis represents the step identification number in the figure.

(30) The input device 28 is operated to specify the pattern matching area on the CAD data, that is, the area corresponding to the secondary electron image to be obtained and the side fixed point P.
This area is designated by half the length f of one side of a square centered on the measurement point P, as shown in FIG. Wiring patterns 40, 41 and 42 are provided in the area in the drawing.
It is included. The data stored in the CAD data storage device 27 is vector data representing the edge of the wiring pattern, and most of these vectors are on the X axis or Y axis of the illustrated XY orthogonal coordinate system with the measurement point P as the origin. It is a parallel vector. In the following processing, the wiring pattern 40 to
Edges e _{1 to} e ₉ parallel to the Y-axis of 42 Pay attention to.

(31) The setting magnification M _S of the secondary electron image is obtained from the value of f and is set in the deflection control device 21. The accuracy of the actual magnification with respect to the set magnification MS of the secondary electron image is ± 2
%. Further, the target position of the stage 11 is obtained from the coordinates of the measurement point P on the CAD data, and this is set in the stage controller 26. The stage controller 26 controls the movement of the stage 11 accordingly. If there is no positioning error of the stage 11, a measurement point P ′ on the sample 12 corresponding to the measurement point P exists on the optical axis of the electron beam irradiation device 10. The positioning accuracy of the stage 11 is
Is about ± 2 μm without feedback control.

(32) From the CAD data storage device 27,
Edges e _{i in} the Y-axis direction of the wiring pattern in the area designated in step 30 (i = 1 to n, n in FIG. 5A)
= 9, and so on), that is, the coordinates of both ends of the edge e _i are extracted.

[0036] (33) determine the range S _i to edge e _i corresponds to on the secondary electron image edge e _i 'is always included. Figure 6
Then, the edge detection ranges S ₁ , S ₆ and S ₇ are indicated by dotted lines.

(34) The electron beam EB is scanned on the sample 12 to obtain a secondary electron image in the SEM image frame memory 23. This secondary electron image is, for example, as shown in FIG.
The wiring patterns 50 to 52 corresponding to the wiring patterns 40 to 42 on the CAD data are included in the secondary electron image. The XY orthogonal coordinate system as shown is also taken on the secondary electron image.
X and Y have a pixel pitch as a unit, and -f'≤X≤f ',
An integer in the range −f ′ ≦ Y ≦ f ′, for example, f ′ =
256. The brightness at the coordinates (X, Y) on the secondary electron image is L (X, Y).

(35) From the brightness L (X, Y), the projection brightness B _i (X), B _i (X) = Σ _iY L (X, Y) on the X axis in the edge detection range S _i is calculated. To do. Here, Σ _iY means the total sum of all Y values in the edge detection range S _i . Further, the edge degree E _i (X) is calculated from the projection brightness B _i (X).

Here, the projected brightness B _i (X) of the uppermost wiring pattern in the secondary electron image changes as shown in FIG. 8A near the edge. The edge degree E _i (X) is expressed as E _i (X) = | B _i (X−q) −B _i (X) | + | B _i (X) −B _i (X + q) | (1) Define in. If the edge width of the projection brightness B _i (X) is 2 W, the preferable value of q for improving the edge detection accuracy is W. In this case, the edge degree E _i (X) is as shown in FIG.
As shown in. The edge width 2W is mainly determined by the size of the electron beam irradiation point. The edge degree E _i (X) does not depend on the right edge or the left edge.

In the wiring pattern below the uppermost layer, usually only the edge portion is detected on the secondary electron image, and the projection brightness B _i (X) is as shown in FIG. 8 (C). The edge degree E _i (X) in this case is as shown in FIG. 8D, and the edge position can be detected.

That is, when the edge degree E _i (X) is defined by the above equation,
The edge of the wiring pattern can be detected or evaluated.

[0042] The X-coordinate of the edge e _i and X _i, the X-axis range of the edge detection range _{S i X i -a i ~X i} + b, the Y-axis range Y
and _{i -c i ~X i + d i} , X i -a i ≦ X ≦ X i + b i becomes X
For j = X−a _i, and for simplification B
_It is expressed as _i (X) = B _i (j) and E (X) = E _i (j).

For example, as shown in FIG. 7A, the projection brightness B ₇ (j) is obtained for the brightness L (X, Y) of the edge detection range S _{7 in} the secondary electron image, and the edge degree E ₇ ( When j) is calculated, it becomes as shown in FIG. Edge degree E ₇
In (j), there are two peaks corresponding to the edges e ₇ 'and e ₈ '.

(37) The pattern matching described later is performed to determine the secondary electron image magnification error ΔM ₀ and the stage positioning error ΔX ₀ .

(38) From the secondary electron image magnification error ΔM ₀ and the stage positioning error ΔX ₀ , the X coordinate of the measurement point P ′ on the secondary electron image corresponding to the measurement point P on the CAD data is obtained. The X coordinate of the measurement point P ′ is ΔX ₀ (1 + ΔM ₀ ). Similarly to the above, the Y coordinate of the measurement point P ′ is obtained. Then, the measurement point P ′ is irradiated with the electron beam EB, and the voltage is measured by a known method. When the Y coordinate of the measurement point P ′ is obtained, if (magnification in the Y-axis direction) / (magnification in the X-axis direction) is detected in advance as a device constant, the magnification increment ΔM described later in the pattern matching process is αΔM. Can be fixed at ₀ .

Next, the pattern matching degree V will be described.

The positioning accuracy of the deflector 17 is ± es / f,
Magnification accuracy of secondary electron image is ± em%, maximum wiring pattern thin width is e1 / f, maximum wiring pattern thin width is e2 /
f, and the maximum wiring pattern edge width is ew / f,
In the case of the right edge of the wiring pattern, a _i = es / f + X _i · em / 100 + e1 / f + ew / f b _i = es / f + X _i · em / 100 + e2 / f + ew / f, and in the case of the left side edge of the wiring pattern, a _i = Es / f + X _i · em / 100 + e2 / f + ew / f b _i = es / f + X _i · em / 100 + e1 / f + ew / f. es, e1, e2 and ew do not depend on the magnification of the secondary electron image.

ΔM, ΔX and ΔW _k are respectively the magnification increment, the shift amount ΔX and the wiring expansion / contraction width (thick or thin) of the kth layer for CAD data, and X _i is X _i (1 + ΔM) +
When changed to ΔX + ΔW _i , j _i = X _i −a _i changes by Δj _i = X _i (1 + ΔM) + ΔX + ΔW _k −a _i − (X _i −a _i ) = X _i ΔM + ΔX + ΔW _k . The k-th layer wiring expansion / contraction width ΔW _k has the same value for the same wiring layer. In pattern matching, as described later, -em / 100≤ΔM≤em / 100, -e
s / f ≦ ΔX ≦ es / f, −e1 / f ≦ ΔW _i ≦ e2 /
The pattern matching degree V is calculated while changing j _i within the range of f, and the magnification increment ΔM and the shift amount ΔX when the pattern matching degree V becomes the maximum are respectively the magnification errors Δ.
It is calculated as M ₀ and stage positioning error ΔX ₀ .

The pattern matching degree V is V = Σ _i Σ _iXY {| L (X−q, Y) −L (X, Y) | + | L (X, Y) −L (X + q, Y) |} e (X, Y) = Σ _i Σ _ix {| B _i (X−q) −B _i (X) | + | B _i (X) −B _i (X + q) |} e (X) = Σ _i Σ _ix E _i (X) e (X) = E ₁ (j ₁ ) + E ₂ (j ₂ ) + ... + E _n (j _n ). Here, sigma _IXY means the sum of all the X, Y set in the edge detection range S _i, Σ _iX means the sum of all the X in the edge detection range S _i, Σ _i
Means the sum of all edge detection ranges S _i . Further, e (X, Y) is 1 if an edge exists at the position (X, Y) on the CAD data, and 0 if it does not exist.
Is. Similarly, e (X) is the position X on the CAD data.
1 if there is an edge at 0, and 0 if it does not exist.

Next, the details of step 37 will be described with reference to FIG.

(371) Initial values are set in variables. That is, 0 is substituted for the wiring expansion / contraction width ΔW, the shift amount ΔX, the magnification increment ΔM, and the maximum pattern matching degree V _max in the formula for calculating the pattern matching degree V. ΔW is the wiring expansion / contraction width ΔW _k of each wiring layer of interest , K = 1 to m are representatively represented.

(372) The pattern matching degree V is calculated.

If (373, 374) V> V _max ,
Secondary electron image magnification error ΔM ₀ and stage positioning error Δ
The magnification increment ΔM and the shift amount ΔX are respectively substituted into X ₀ .

(375) The wiring expansion / contraction width ΔW is updated. This update is performed on a pixel-by-pixel basis.

(376) If the updating of the wiring expansion / contraction width ΔW has not been completed, the process returns to step 372, and if it has been completed, the process proceeds to the next step 377.

(377) The shift amount ΔX is updated. This update is performed on a pixel-by-pixel basis.

(378) If the updating of the shift amount ΔX is not completed, the process returns to the step 372, and if it is completed, the process proceeds to the next step 379.

(379) The magnification increment ΔM is updated. This update is performed, for example, so that the pixel at the position (f, 0) in FIG. That is,
Let ΔM = 1 / f.

(37A) If the updating of the wiring expansion / contraction width ΔW is not completed, the process returns to step 372, and if completed, the pattern matching process is completed.

As described above, the secondary electron image magnification error Δ
M₀And stage positioning error ΔX ₀Is required.

(2) Second Embodiment In the first embodiment described above, only the case where the wiring pattern is parallel to the X-axis or the Y-axis has been described, but the present invention uses an oblique pattern which is not parallel to the X-axis and the Y-axis. It can also be applied, which will be described next.

As shown in FIG. 9F, it is assumed that the wiring pattern 43 has the edge e _A on the CAD data, and the detection range is the range S _A shown by the dotted line. When this edge detection range S _A is applied to the secondary electron image, FIG.
As shown in. When the projection brightness distribution is calculated by calculating the sum of the brightness L along a straight line parallel to the hypotenuse of the edge detection range S _A , as shown in FIG.
W becomes relatively wide and pattern matching becomes inaccurate. The oblique projection brightness in this case is as shown in FIG. The widening of the edge width 2W is due to the magnification error.

Therefore, in the second embodiment, as shown in FIG. 9B, the edge detection range S _{A is set} to, for example, the edge detection range S
It is divided into three parts, _A1 , S _A2 and S _A3 . Edge detection range S
Oblique projection luminance B _A1 in _A1 (j) becomes as shown in FIG. 10 (C), the edge width in the case of 2W / 3 and FIG. 10 (A) 1
/ 3. The edge degree E _A1 (j) in this case is shown in FIG.
As shown in FIG. 10D, the peak becomes sharper than in the case of FIG. As for the pattern matching degree V, the edge degree is obtained for the edge detection range divided in this way, and the sum is taken in the same manner as above. Further, the enlargement direction of the diagonal pattern is an arrow direction perpendicular to the edge straight line before enlargement, as shown in FIG. The other points are the same as those of the first embodiment.

[0064]

As described above, according to the pattern matching apparatus of the present invention, the edge degree E (X) is obtained from the brightness distribution B (X) obtained by projecting the secondary electron image on the axis X, and this edge is calculated. The pattern matching degree V between the degree E (X) and the one-dimensional edge position X _i , i = 1 to n on the CAD data is obtained, and the edge position X _i , i = 1 to n is calculated as the secondary electron image magnification accuracy, Since the secondary electron image magnification error and the sample positioning error which maximize the pattern matching degree when changed based on the sample positioning accuracy and the wiring width scaling, the wiring pattern on the CAD data and the wiring pattern on the secondary electron image are obtained. It has an excellent effect that pattern matching with and can be performed easily, accurately and in a short time, and largely contributes to improvement of operability of the electron beam apparatus.

According to the first aspect of the present invention, the pattern matching degree V is V = E (X ₁ ) + E (X ₂ ) + ...
Since it is calculated by + E (X _n ), the pattern matching degree V
It is possible to evaluate accurately and easily.

According to the second aspect of the present invention, the edge degree E
(X) becomes E (X) = | B (X−q) −B (X) | + | B
Since it is calculated by (X) -B (X + q) |, the edge of the wiring pattern can be correctly evaluated regardless of the wiring layer, the right edge, and the left edge.

According to the third aspect of the present invention, the edge position X
_i , i = 1 to n range X _i −a _i ≦ X _i ≦ X _i + b _{i Since} the projection brightness B (X) is calculated only, the data to be processed becomes the minimum necessary, and the data processing is performed in a shorter time. There is an effect that can be done.

According to the fourth aspect of the present invention, the edge straight line of the diagonal wiring pattern which is non-parallel to both axes of the orthogonal coordinate system on the CAD data is also detected, and the straight line on the secondary electron image parallel to the edge straight line is detected. Then, the distribution of the oblique projection brightness obtained by adding the brightness of the secondary electron image along is also found, the edge degree is detected from the oblique projection brightness distribution, and the degree of correlation between the edge straight line and the edge degree is also found to obtain the pattern. The edge straight line is added to the matching degree V, and this edge straight line is also changed based on the secondary electron image magnification accuracy, the sample positioning accuracy, and the wiring width expansion / contraction to obtain the secondary electron image magnification error and the sample that maximize the pattern matching degree V. Since the positioning error is obtained, it is possible to perform the pattern matching process on the diagonal wiring pattern similarly to other wiring patterns.

[Brief description of drawings]

FIG. 1 is a principle configuration diagram of a pattern matching device according to the present invention.

FIG. 2 is a main part configuration diagram of an electron beam tester to which the present invention is applied.

FIG. 3 is a flowchart showing a voltage measurement procedure.

FIG. 4 is a flowchart showing a pattern matching procedure.

FIG. 5 is a diagram showing a designated area on CAD data and a secondary electron image obtained based on the designated area.

FIG. 6 is an edge detection range S obtained based on CAD data.
_It is a figure which shows _i .

FIG. 7 is a diagram showing an edge detection range and a distribution of edge degrees obtained from luminance data in the range.

FIG. 8 is an explanatory diagram for obtaining an edge degree from projection brightness.

FIG. 9 is a diagram showing an edge detection range of an oblique pattern obtained based on CAD data.

FIG. 10 is a distribution diagram of projected brightness and edge degree of a diagonal pattern.

FIG. 11 is an enlarged explanatory view of a diagonal pattern.

[Explanation of symbols]

10 electron beam irradiation device 11 Stage 12 sample 20 secondary electron detector 40 to 43, wiring patterns - down _{e 1 ~e 9, e A,} e 1 '~e 9', e A ' edge S ₁ to S _9, S _A , S _A1 ~ S _A3 Edge detection range P, P'Measurement point

Claims (5)

[Claims] 1. A pattern matching device for performing pattern matching between a wiring pattern of a secondary electron image obtained by scanning an electron beam on a sample (2) with an electron beam device (1) and a wiring pattern on CAD data. At edge position detecting means (3) for detecting the position X _i , i = 1 to n of the wiring pattern edge parallel to one axis Y of the orthogonal coordinate system on the CAD data, and the secondary electron image is stored. Secondary electron image storage means (4)
And a projection brightness B obtained by adding the brightness of the secondary electron image along a straight line parallel to one axis Y of the orthogonal coordinate system on the secondary electron image.
Projection brightness distribution creating means (5) for obtaining (X), and edge degree E of the wiring pattern from the projection brightness B (X)
An edge degree detecting means (6) for detecting (X) and a pattern matching degree calculation for obtaining a degree of correlation between the edge positions X _i , i = 1 to n and the edge degree E (X) as a pattern matching degree V. Means (7) and edge positions X _i , i of the wiring pattern on the CAD data
= 1 to n are changed within the range of X _i −a _i ≦ X _i ≦ X _i + b _i based on the secondary electron image magnification accuracy, sample positioning accuracy, and wiring width expansion / contraction, the pattern matching degree becomes maximum. An error detection means (8) for obtaining a secondary electron image magnification error and a sample positioning error, and a pattern matching device characterized by the above. 2. The pattern matching degree calculation means (7) calculates the pattern matching degree V as V = E.
_2. The pattern matching device according to claim 1, wherein the calculation is performed by (X ₁ ) + E (X ₂ ) + ... + E (X _n ). 3. The edge degree detecting means (6) calculates the edge degree E (X) as E (X) = | B (X−q) −B (X).
The pattern matching apparatus according to claim 2, wherein the calculation is performed by | + | B (X) -B (X + q) |. 4. The projection brightness distribution creating means (5) has a range X _i −a _i ≦ X _i ≦ X of the edge position X _i , i = 1 to n.
The pattern matching apparatus according to claim 1, wherein the projection brightness B (X) of only _i + b _i is obtained. 5. The edge position detecting means (3) also detects an edge straight line of an oblique wiring pattern which is non-parallel to both axes of the Cartesian coordinate system on the CAD data, and the projection luminance distribution creating means (5). Also obtains a distribution of oblique projection brightness obtained by adding the brightness of the secondary electron image along a straight line on the secondary electron image parallel to the edge straight line, and the edge degree detecting means (6) The degree of edge is also detected from the projected luminance distribution, the pattern matching degree calculating means (7) also obtains the degree of correlation between the edge straight line and the edge degree, and adds the degree to the pattern matching degree V, The error detecting means (8) also changes the edge straight line of the diagonal wiring pattern on the CAD data based on the secondary electron image magnification accuracy, the sample (2) positioning accuracy, and the wiring width expansion / contraction to obtain the pattern matching degree. V is the maximum Pattern matching apparatus according to claim 1, wherein the determination of the secondary electron image magnification error and sample positioning error that.