JPH0214372A - Method for preparing cad data for artwork graphic - Google Patents

Abstract

PURPOSE:To easily obtain the CAD data of an artwork graphic by previously registering both aperture shape data and wiring width data to indicate the line width of a wiring graphic, separating the data into respective graphics, separating the data into wiring CAD data, land graphic CAD data, and non-land graphic CAD data, and creating the CAD data. CONSTITUTION:An operator inputs the values of aperture graphic data for a land, the wiring width data, the maximum aperture area FAmax, the minimum aperture area FAmin, etc. The artwork graphic is read by an image pickup means, picture data are converted into dot data as binary data, and the dot data are converted into outline side vector data CBD1 in a work station. As the result of the processing, for a graphic F1, the wiring CAD data to indicate a wiring graphic Fw1 and the land graphic CAD data to indicate a land graphic respectively are created. Next, fine-linearizing processing using the wiring width is executed, all the above-mentioned data are detected as wiring information, and the wiring CAD data are created.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a method for creating CAD data for artwork figures used in designing and manufacturing printed wiring boards.

(Prior Art) Conventionally, in the story and production of printed wiring boards, artwork figures have been created by manually creating the figures of the wiring patterns on a predetermined film or the like.

On the other hand, with recent advances in CombiCoke technology, it has become possible to create artwork shapes using a CAD system, and it has become possible to modify artwork shapes in response to changes in the wiring pattern design. .

(Problem to be Solved by the Invention) However, the artwork figure (b) remains, for example, as an original image printed on film, and there are cases where there is no CAD data to be input into the CAD system. In such a case, there is a problem in that the CAD system cannot be used to change the design of the artwork figure.

Additionally, it is possible to create CAD data by using a digitizer by an operator to input graphic information based on the original artwork graphic, but this graphic information must be entered manually (because However, there was a problem in that it required a great deal of time and effort.

(Purpose of the Invention) This invention was made to solve the above-mentioned problems in the prior art, and it provides an artwork that can easily (9) create CAD data based on the original drawing of the artwork figure. The object of the present invention is to provide a method for creating CAD data of figures.

(Means for Achieving the Object) In order to achieve the above object, the present invention provides an artwork shape having a plurality of land shapes having a predetermined shape and a wiring shape having a predetermined line width that interconnects these land shapes. CAD
In the method for creating data, (a) aperture shape data indicating the shape of the land figure and line width data indicating the line width of the wiring figure are recorded in advance, and (b) the artwork figure is read. After creating contour side vector data of the artwork figure based on the image data obtained in step 17, (c) By thinning the side vector data and (d) extracting a vector portion having a width smaller than a predetermined reference width as a wiring vector in this thinned contour side vector data, the position data of the wiring figure and the above While obtaining wiring CAD data having line width data,
Non-wire vector data is obtained by removing the wire vector from the contour side vector data, and (e) pattern matching is performed between the non-wire vector data and the aperture shape data to combine the non-wire vector data with the aperture shape data and the non-aperture shape data. ([) Land figure CAD data having the land figure position data and the aperture shape data are obtained, and non-land figure CAD data is obtained from the contour side vector data of the non-aperture shape data. demand.

(Operation) The artwork figure includes a land figure of a predetermined shape, a wiring figure of a predetermined line width, and other figures.

Regarding the land figure, land figure CAD data having position data of the land figure and aperture shape data indicating a predetermined shape of the land figure is obtained.

For a wiring diagram, a wiring CA having position data of the wiring diagram and line width data indicating a predetermined line width of the wiring diagram.
D data is obtained.

For other figures, non-land figure CAD data is extracted from their contour side vector data.

As a result, as the CAD data of the artwork figure, land figure CADF data, wiring CAD data, and non-land figure CAD data are obtained separately.

(Example) A. Schematic configuration of artwork figure input device FIG. 1 shows a schematic configuration of an artwork figure input double-printing device as a device for creating CAD data of artwork figures by applying an embodiment of the present invention. It is a diagram.

In the figure, the artwork graphic input device ff1AP is connected to workstation 1. External input means 2. Storage means 4. Specific power stage 5. and CAD data conversion means 6. Further, a CRT 3 is connected to the external input means 2.

The artwork AW on which the artwork figure F is drawn is repeatedly scanned along the X direction at every predetermined width in the X direction by an imaging means 11 composed of a COD or the like, and the image data of the artwork figure F is read. It will be done. The artwork figure F is a land figure F having a predetermined shape (aperture shape) such as a circle or square, or a wiring figure F8 drawn with a standardized predetermined line width. and a filled-in figure FR having an arbitrary shape. This artwork figure F is used for manufacturing a printed wiring board, the land figure F corresponds to the position of a bin of an IG package, and the wiring figures F and 4 correspond to electrical wiring. Hereinafter, when these are collectively referred to, they will simply be referred to as "figure F."

The density data fx for each pixel read by the imaging means 11 is binarized by the scanner section 12 and provided to the workstation 1 as dot data Dx for each pixel. Workstation 1 is dot data D
Based on x, closed-loop contour force vector data (hereinafter simply referred to as "contour side vector data") CBD that represents the contour sides of figure F as closed-loop vectors is created. This contour side vector data CBD is given to the CAD data converting means 6, and finally the land figure CAD data D1. The wiring CAD data Dw and non-land figure CAD data DR are converted into CAD data multiplied by +111.

The CAD data converting means 6 is constituted by a computer, and has each function realizing means 61 to 65, which will be described in detail later.

Land figure C obtained by the CAD data conversion means 6
AD data D[, distribution mcΔD data D.

and the non-land graphic CAD data DR are transmitted to the workstation 1 and output to a magnetic tape or the like by the output means 5. Furthermore, wiring width data, aperture figure data, etc. are input in advance by the operator from the external input means 2 and are registered in the CAD data conversion means 6. The CRT 3 connected to the external input means 2 reproduces the figure F based on the above-mentioned CAD data D, D, DR, and is used by the operator to check and correct the CAD data D, D, DR. Ru. Additionally, various other data are displayed as necessary. The storage means 4 is for storing various data necessary for the work of the workstation 1.

B. Details of processing in the CAD data converting means 6. 10 Contour side vectors 11. C [28 figures for 3D cows to 2nd D- are density data fx. Contour side vector data CBD is created based on the density data fx. FIG.

FIG. 2A shows the density data f)! obtained by reading the artwork AW. , an example is illustrated. The density data fx is obtained as an analog value corresponding to 11 degrees of the artwork AW for each pixel Px whose minus side is several tens of μm. However, in FIG. 2A, the pixel Px having a large value of the density data f is indicated by diagonal lines, and the figure F is composed of all the pixels indicated by the diagonal lines.

Such density data fx is given from the imaging means 11 to the scanner section 12, and is binarized into dot data Dx,
is converted to The flow of this conversion process is shown in FIG.

First, in order to remove noise components, smoothing is performed using a method such as a moving average method or a median filter. Next, threshold processing is applied to the smoothed density data,
Perform binarization. Further, by enlarging/reducing the binarized figure, minute protrusions and defects in the binarized figure are removed. In this way, noise components are removed and 2
Valued dot data Dxy is obtained.

FIG. 2B shows dot data Dx. In the figure, the value of dot data Dx is 1" for pixels inside figure F, and "O" for other pixels. Such dot data Dxv is transferred from scanner section 12 to workstation 1. Given.

Based on this dot data D, the workstation 1 first sets all the values of the dot data Dx in the xy pixels other than the outline pixels of the figure F to "0". For this process, a 3×3 matrix M shown in FIG. 4 is used. However, “X” is II II II
However, it is shown that b is good at °O''.For each area consisting of 3 x 3 pixels extracted from Fig. 2B, the dot data Dxy and the value of the 3 x 3 matrix M are compared,
If the dot data D corresponding to the cross-shaped portion M of the 3×3 matrix M is all 1”, then 3×371−
It is determined that it matches the ricks M. In the area that matches the 3x3 matrix M, the dot data Dx corresponding to the center part M of the 3x3 matrix M is "1".
” to “0” and converted dot data Dx
, ′ is obtained. At this time, the dot data D with B is retained as is, and the converted dot data D' is stored separately in the storage means 4.

xy On the other hand, in a region that does not match the 3×3 matrix M, the dot data Dx corresponding to the center portion M is maintained as is and is taken as the converted dot data D'. In this way, the dot data Dx,', in which only the outline of the figure F as shown by xy in FIG. 2C is 1'', is (q).

Next, the contour vector Cr5 (81 to B8 in FIG. 2D) is created by tracing the axis of °1'' of this dot data [ ] x, l. Note that the contour vector Cr5 of the leap loop Individual subvectors B −8 constituting CB
8 will be simply referred to as a "vector" hereinafter. In this example,
The right side of each vector 81 to B8 is an area where the value of the dot data D is "1" (hereinafter referred to as a "black area").The left side is the dot data Dx, and (7) the area where the value is "0" is the i! 'i area (hereinafter "
"white area". ) is supported. That is, a clockwise contour vector indicates that the interior thereof is a black region, and conversely, a counterclockwise contour vector indicates that the interior thereof is a white region.

FIG. 5 shows an example of creating such a vector B (B to B8) by the polygonal line approximation method.

The contour portion shown as a series of pixels whose dot data Dx,' is "1" in FIG. 2C is shown by a curve S in FIG. Since the pixel Px is usually very small, the contour can be regarded as such a curve. In the polygonal line approximation method, the maximum distance d between the vector B and the contour S. (not shown) is specified in advance from the external input means 2 or the like.

The distance d between the vector B connecting two points on the contour S and the contour S is the maximum value d. Under the following conditions, connect two points as far apart as possible from each other and define vector B by connecting /υ.

Vector B shown in FIG. 2D obtained in this way
-88 data is leap loop contour side vector data (
Hereinafter, this will be referred to as "contour side vector]-le data." ) are summarized as CBD. The contour vector data CBD includes one closed loop vector management information CBM and each vector 81.
Vector data BD1-B for each of ~B8
Consists of D8. Leap loop vector management information CBM
includes the number of vectors forming the closed loop (=8) and connection information with other idle loop vector data CBD. In addition, each vector data BD, to BD8 is vector information B1 to BI8 and vector concatenation information BC1 to BC.
8, respectively. Vector information BI ~B
I8 is composed of each vector 18 IT point coordinates (x, y). In addition, [E vector concatenation information BC1 to B08 is for each vector B to
This is data indicating which other vectors B1 to B8 each of B is connected to.

The connection information between the 11 loop vector data included in the open loop vector management information CBM is information indicating the bearing relationship between the inner peripheral contour and the outer peripheral contour of each figure.

Therefore, the contour vectors CB1 to CB5 generated from the figure F in FIG. 7A are assumed to have a connected relationship as shown in FIG. 7B. Figure F in Fig. 8Δ. F2
As shown in FIG. 8B, the connection relationship indicated by the broken line may be severed later.

B-2. Detection of isolated fish head Contour side vector data CB created as above
D is the isolated point area detection means 6 from the workstation 1.
1 is given. FIG. 9 is a conceptual diagram showing an isolated point area detection processing method. In the figure, the figure F is represented by a clockwise contour side vector CB, and the size of the figure F is represented by a width ΔX1 in the X direction and a width Δy in the X direction.

The figure also shows the maximum aberration i! specified in advance by the operator from the external input means 2! Area FAIllax is shown. The maximum aperture area ll1FA18x is specified in advance as an area slightly larger than any of a plurality of seven perforation figures F standardized as described later.

When the figure F is smaller than the maximum aperture area FA□8, the figure F is determined to be an isolated point area, and it is determined whether it matches any of the plurality of aperture figures F. The maximum aperture area FA□8 is
width in the direction (hereinafter referred to as "x direction maximum aperture area").

) ΔXH and the width in the X direction (hereinafter referred to as "X-direction maximum aperture region") Δy8. Figure F represented by contour side vector CB
If it satisfies the following equations (1) and (2), the figure F is considered to be in an isolated point region.

ΔX < ΔXH... (1) Δ y × Δ 'WH... (2) Here, Δx: Maximum aperture in the x direction + 7 areas Δy: Maximum aperture 1j area in the y direction ΔX = Width in the X direction of figure F Δy: X-direction width of figure F FIG. 10A is a conceptual diagram showing an example of isolated point area detection processing when one figure F1 includes another figure F2. As shown in the figure, the maximum aperture area FA
is smaller than the figure F1 and larger than the figure F2. Therefore, the figure F2 is determined to be an isolated point area. The connection relationship between the contour side vector data corresponding to FIG. 10A is as shown in FIG. 108, and when the figure F2 is determined to be an isolated point area, the contour side vector data CBD and CBD2 representing the figure F1 are as follows. It is transmitted from the isolated point area horizontal means 61 to the thinning processing means 63,
Wiring information detection processing, which will be described later, is performed. Also, figure F
The contour side vector data CBD3 representing 2 is transmitted to the aperture recognition means 62.

Furthermore, as can be seen from the above explanation, the maximum aperture I
Only the clockwise contour side vectors CB and CB3 are compared with the /area FAlax.

This is because the shape of the outermost periphery of the figure in the black area is always represented by the contour side vector around time i1.

The figure F is also compared with a minimum aperture area FA1° (not shown). Minimum aperture area FA,. is specified in advance as an area smaller than any of the standardized aperture shapes.

Therefore, the figure F is the minimum aberration lν area FA. If the error information E is smaller than
1 to the workstation 1, and is further displayed on the CRT 3 via the external input means 2. This process will be described further later.

In this way, the contour side vector data of the figure smaller than the maximum aperture area FAIIlax and larger than the minimum aperture area FAsin is transferred from the isolated point area detection means 61 to the aperture recognition means 62.

B-3. Wiring information detection processing FIG. 11 is a conceptual diagram showing the procedure of wiring information detection processing for figure F. In the figure, figure F is a square 2
It consists of two land figures F and F and a wiring figure F connecting them.

The maximum aperture area F in the isolated point area detection means 61
The contour side vector data CBD of the figure F that is determined to be larger than Alax is transferred to the thinning processing means 63, and wiring information is detected by the steps described below.

Step C: Separation thinning process (Figure 11 (a) → (b)
)) Step B: Detection of wiring information and separation processing of wiring figures and non-wiring figures (Fig. 11 (b) → (c)) Step C:
Thinning reverse processing II for separated non-wiring figures (first
1(c)→(d)) FIG. 12 is a conceptual diagram showing the method of thinning in further detail. In FIG. 12A, the contour side vector CB of figure F is sequentially connected vectors 81 to 81.
It is composed of B14. Further, the figure F is composed of two land figures FA, FA, 2, and a wiring figure F that connects these to each other. Also, wiring diagram F
. The wiring width t is approximately equal to a predetermined reference wiring width width inputted in advance by the operator from the external input means 2. The thinning is performed by moving each of the vectors 81 to B14 in FIG. 12A to the right side of the advancing direction by 1/2 of the base wiring width T, thereby reducing the figure F as a whole.

Figure 12B shows the figure F* that has been thinned in this way, and for comparison, the figure before thinning is shown with a broken line.Generally, the coordinates of the end point of vector B are vector Bi+1 , and these coordinates are indicated by mutual connection points cp, (+=1 to 14).Therefore, thinning as shown in Fig. 128 means that the thinned vector New connection point CP of B1*~B14'
, is equivalent to finding ``.

Figure 13 shows the connection point CP of vectors Bi and Bi+1.
From the coordinates (x, y □), the coordinates of the connection point cp," between the thinned berittle B, ゝ and 8.8 + 14
1 Shows how to find 1(x, yl) 1'

The connection points cp after thinning are solid 1 to B, and Bi+
Exists on a straight line that bisects the angle 2α formed by 1, and
The vectors B and Bi+1 are located at a position where the vertical distance from each straight line is 1/2 of the wiring width T. Such a position is the 13th
As shown in the figure, point C on the right side of vector B・, Bi+1
There are two points, P,'' and the left point CP,'. In this example, the vectors B,,B,+.

The right side with respect to the direction of movement is defined as a black area. Therefore, the connection point after thinning is the vector Bi
, J+1 is selected. The coordinates (xl, y1') of this new connection point cp,'' are calculated as follows, for example.

The distance 1 between the connecting points cp, , cp,'' before and after thinning has the following relationship with the reference wiring width T, the angle 2α formed by the vector B· and Bi+1.

sin α Also, the coordinates (x, yl1 1) of the new connection point cp," are the difference in the X! mark, y coordinates +, y+ from the coordinates (x, yo) of the original connection point cp. It can be expressed as follows.

x = xo+x yl = yo+y On the other hand, the connection points cp, ,cp If the angle between is expressed by the following equation.

...(4) ...(5) The difference between the straight line connecting 8 and the X axis above x", y" sinα nα Also, from the contour side vector arta,
, Nona 1 angle 01. and the angle θi+1 between the X axis and the vector B. From these angles θ, , θih1, the above angles α and β can be calculated as E1 as follows.

α=1/2(π-(θ,-θi+1))=・π/2-(
θ, -θ, )/2 ... (8) 11+1 β-α-(π-θi) = -π/2+ (θ・-to Oi, 1
) /2 ... (9) (6) < i. From the formula 9), the difference can be written as:

cosθb 03Ob where θ − (θ → −θ, , )/2θ −<0
-0. , V2 The above calculation method is just an example,
Needless to say, there are various other calculation methods. If new connection points CP1 to CP14* are determined based on the above formulas (4), (5), (10), and (11) and these are connected, a thinned figure will be created as shown in Figure 12B. Vectors B1* to B14* are obtained. 128th
In the figure, for convenience of illustration, the partial vector 84*11* is drawn as being separated from the partial vector B4, but the original vector B4. When the mutual distance t of B11 is equal to the reference wiring width T, these vectors 84* and 811* overlap each other.

Note that when the above-mentioned thinning is performed, the contour edge of the clockwise contour edge vector is reduced, and conversely, the contour edge of the counterclockwise contour edge vector is expanded.

Therefore, as shown in figure F1 in FIG. 10A, for example, the clockwise contour blade vector CB1 and the counterclockwise contour blade vector C
It can be seen that the width of the black area of the figure constituted by B2 is also narrowed by the reference wiring width T.

B-3b. Processing In this way, the contour edge vector data CBD'' of the thinned figure F is transmitted to the wiring information detection means 64. FIG. 14 shows the thinned figure F. FIG. 3 is a conceptual diagram showing a method of detecting wiring information from *.

Whether or not this figure F* includes the wiring figure F'' is detected as follows.Now, regarding the vector B1*, the area in the X coordinate direction between its starting point CP14'' and its ending point CP Bi is detected as follows. xf! It is defined as four areas Rx1, and the area in the y coordinate direction is defined as y area Ry1. The starting point CP14'
If the coordinates of is (X14.!l'14) and the coordinates of the end point cp1'' are (x, y), then the X area Rx1 is (x14→
x1), and the y area Ry is (y14→y1
) area.

Next, this X region Rx and yfR region Ry1 are compared with the respective X regions and X regions of the other vectors 821 to B141. If there is a vector that forms a wiring diagram together with vector B1*, the X area and X area of that vector and the X area Rx and y area Ry of vector B11.
There should be parts that overlap with 1. As can be seen from FIG. 14, it can be seen that for vector B, there is no such vector and no wiring diagram is formed.

On the other hand, even when the above-mentioned pattern exists, there are times when it forms a wiring diagram, and times when it does not. The distinction is made as follows. For example, the X region Rx4°y region Ry4 of the vector B4'' almost overlaps with the X region Rx11°yfft region R''l/11 of the vector B11*.

At this time, the starting point CP3' of vector B4* and vector B
The mutual distance between the end point CP11* of vector B4* and the starting point CP1o* of vector B4* is shorter than the threshold Δtwax, and the distance between the end point CP11* of vector B4* and the starting point CP1o* of vector B4* is shorter than the threshold value Δtwax. When shorter, the vectors 84' and 8111 are the wiring diagram F6
It is judged that it forms a At this time, the wiring diagram Fw
The position of * is given by the center line of vectors B41 and 811'. Specifically, the starting point P8 of the wiring diagram F1,
The coordinates (X, V) of (not shown) and the coordinates (x, y) of the end point PW [(not shown) are given as follows.

Death correction xWl” (x3 +x11)/2・-(12a)V
= ('W 3 + V 11) /2 ・
・・(12b)−■
・(13a)y = (y4+ylo)/2
...(13b) - [However, connection point CP," (i = 3.4, 10.11)
The coordinates of are respectively (×・, Vi).

The starting points P, 4□ and the ending point PWE of this wiring diagram F1 are the first
These are the starting and ending points of the wiring diagram FW before thinning shown in Figure 2A. Also, wiring diagram F. As the data indicating the wiring width, the standard wiring width T used for thinning is used instead of the actual wiring width t. In other words, the wiring diagram FW
As data representing the wiring C, wiring C AD data, which is composed of the starting point coordinates (xy), the ending point coordinates -Wl (x, y), and the wiring width T, is obtained.

In this way, the wiring figure FW shown in FIG. 12A is detected as a wiring having the standard wiring width T lax when the difference between the wiring width t and the reference wiring width T is smaller than the threshold value Δt. . This threshold Δ
It is best to set t to a length corresponding to several pixels of image data, with lax depending on the accuracy of reading the wiring diagram.

Note that when the wiring width t is 6 smaller than the standard wiring width f,
It is also considered that the mutual positions of the vectors 84' and B11* are interchanged, but even in such a case, the phase n distance is equal to the threshold value Δ
When it is smaller than t, distribution I! Detected as a laX line figure.

Fig. 15Δ and Fig. 158 are two diagrams composing the wiring diagram.
1' is a conceptual diagram showing a method for detecting wiring information when two vectors do not completely match. FIG.

In FIG. 15A, two vectors B after aFA conversion
and 8. The distance between them is sufficiently short and they form a wiring diagram. Also, the X region RXb of vector Bb, the 'i' region R
yb is the X region [Rx, y region R
Included in y8. However, starting point C of vector B
The distance between P81 and the end point CP of vector B is h2 from the threshold Δt.
111aX is also much larger. The same is true for the distance between the end point CPa2 of the vector B8 and the start point CPb1 of the vector Bb. In such a case, in order to detect the wiring diagram, a point PP82 having the same X coordinate as the starting point CP of the vector B is found on the vector Ba. Then, when the distance between these points CP, 1 and point PPa2 is smaller than a predetermined threshold value Δtymax shown in the figure, it can be determined that a wiring figure exists fgi. This predetermined Li1
The l value Δtylax is a value set in advance as a length of several pixels, and the same value as the thresholds Δt1 and lax described above may be used. If it is determined that a wiring diagram exists, a point PP having the same X coordinate as the end point CPb2 of vector B is found on vector Ba. Then, the midpoint between the points PPa□ and CP and the midpoint between the points PP and CPb1 are determined as the positions of both ends of the wiring diagram, b2 and a2, respectively.

FIG. 15B is a diagram illustrating an example of a diagram including wiring diagrams.

In the figure, vector B. × territory IJ! iRx. ,yf
A part of each of fi[Ryo overlaps with a part of each of the X region RX(, V region Ryf) of vector B[. In a case like this figure, first, vector B. Above,
A point PP having the same X coordinate as the end point CP of vector B
o1 is determined, and the midpoint between the point PPo1 and the end point CPf2 is set as one end of the wiring diagram. Also, a point P on vector Ba that has the same X coordinate as the end point CPo2 of vector B
Pfl is determined, and the midpoint between the point PPH and the end point CPo2 is set as the other end of the wiring diagram. In this way, the positional coordinates of both ends of the wiring diagram portion can be accurately obtained.

J3 In the above example, the point PP at the X coordinate around the starting point or ending point of one vector is found on the other vector. However, when finding these points PP, the y coordinate may be used as a reference instead of the x coordinate. Alternatively, a straight line passing through the starting point or ending point of one of the vectors and perpendicularly intersecting the other vector may be found, and the intersection of that straight line and the other vector may be used instead of the above-mentioned point PP. Furthermore, when the wiring diagram [, 4 faces the X-axis direction or the Y-axis direction, only one of the X region or the Y region overlaps, and the other region exists at a distance aX smaller than the threshold value Δt described above. I will do it. In such a case, it is determined whether or not it is a wiring diagram based on the overlap of either the X area or the Y area, the similarity of the angles of the respective vectors B1, and the size of the distance between them. All you have to do is determine.

As described above, the portion of the wiring figure F1 is detected from the thinned figure F*, and the wiring CΔD data D is obtained.

As shown in FIG. 11(b) → (cl), the thinned figure F8 is separated into wiring CAD data D and contour side vector data of other figures (land figures F11*, L2'' in the figure). Among these, wiring CAD data Dw
is transmitted from the wiring information detection means 61 to the workstation 1.

On the other hand, the island normal contour side vector data CBD is obtained by removing the wiring figure from the thinned contour side vector data CBDI.
" is transmitted from the wiring information detection means 64 to the thinning inverse processing means 6 and 5. In the thinning inverse processing, the contour sides of the figure are gradually This is a process of enlarging the kite V by 1/2 of the radius T. As a result, the first
As shown in Figure 1(d), from the original figure F to the wiring figure F
, the contour side vector [1L2 CB CB82 is obtained for the land figure F , F with , removed.

al. In addition, figure F obtained by thinning inverse processing, 1. The contour side vectors CB, CB, 2 of FL2 are the same as those of figure FL1. Interpolation vectors B1+ and 82+# are placed at the positions where FL2 was connected to wiring diagrams F and l, respectively.
As a result, the contour edge vectors CB, CB
, 2 are closed loops [1].

Then, these contour edge vectors]・ruCB,1. CB, 2
Contour side vector data CBD representing 1. CBDL2
is from the thinning inverse processing means 65 to the isolated point area detection means 61
transmitted to. The isolated point area detection means 61 detects that the figures F and F represented by the transmitted contour side vector data C B D and 1° C.D.
It is determined by the method described above whether or not the LI L2 position is in the standing point area, and if it is in the isolated point area, the data is transmitted to the aperture recognition means 62.

Further, when the figures F 1 and F 2 are not in the isolated point area, the thinning processing means 63 returns data CBDL1 . Transmit CBD[2 and perform the above processing.

Note that the artwork figure generally has N types of wiring figures having different standard wiring widths T from each other. In this case, the reference wiring width T, (i=
1 to N) is selected and thinning processing is performed, and wiring information detection processing is performed for each reference wiring width T.

If it is determined that the figure is neither a wiring figure nor a figure within an isolated point area, the figure is determined to be a filled-in figure. The outline side vector data of the filled figure is directly used as non-land figure CAD data DR,
It is transmitted from the thinning processing means 63 to the workstation 1.

B-4. Aperture Recognition Processing The figure determined to be in the isolated point area by the isolated point area detection means 61 is examined by the aperture recognition means 62 to see whether it matches several types of standardized aperture figures.

FIG. 16 is a conceptual diagram showing an example of a land figure after the wiring figure has been removed. When the wiring width of the wiring figure is narrow, the remaining land figures F 1 and F 1 are 1L1 as shown in FIG. 16(e), as shown in FIGS. 16(a) and (b).
[2] They have almost the same shape as the 7J2 scaled aperture figures F and F shown in (f), respectively. However, when the wiring width of the removed wiring diagram is thick as shown in FIGS. 16(c) and (d), A1 82 is the land diagram F.
, F are considerably different in shape from the corresponding aperture figure [3[4 F , F ].
I A2. Matching between land shape and aperture shape is
A land figure and an aperture figure represented by dot data for each pixel are compared, and a determination is made based on the proportion of pixels that match each other. Therefore, FIGS. 16(c) and (d)
The land figures F and 3°F as shown in are paired with the aperture figures F and F, respectively.
There is a possibility that it will be determined that it does not comply with AI A2.

Furthermore, since pattern matching using dot data requires a lot of processing time, there is a problem that the processing time will be long if a land figure and a large number of aperture figures are to be compared by pattern matching. Therefore, in order to accurately and quickly find the averge V figure corresponding to the land figure, the following measures have been taken.

First, before performing pattern matching, aperture figures to be compared are selected as follows.

Most of the land shapes (aperture shapes) used in artwork shapes are squares, hexagons, or circles, and none have protrusions. Therefore, an aperture figure is selected using two land figures using two types of interpolation vectors: a linear interpolation vector and a circular interpolation vector.

FIG. 17 is a conceptual diagram showing an example of a land figure using these two types of interpolation vectors. Figure 17 (a-1)
, (a-2) shows the land figures F''La2 corresponding to the square aperture figure FAs, respectively.The land figures F1,1 are linear corrections connecting the two connection points CP CP82 of the part from which the wiring figure has been removed. has a vector BLa+ between a1 and a1.On the other hand, the land figure F
, 82 are the two connection points CPa1. A circular interpolation vector B is a circular interpolation vector B in which a plurality of small vectors connect the circular arcs on the left side with respect to the traveling direction of the vector among the circular arcs whose diameters are at both ends of CP,2. It has a+. In addition, for each land figure in FIG. 17, the outline of the corresponding aperture figure is shown by a broken line for comparison.

Here, the length of a plurality of vectors constituting the contour side vector of each land figure is integrated to obtain the peripheral length of each land figure. As can be seen by comparing both FIGS. 17 (a-1) and (a-2), the perimeter of the aperture figure FAS is the land figure F with linear interpolation vectors B and a+.
, 81, and the circular interpolation vector B. 8
It is shorter than the peripheral length of the land figure FLa2 having +. This is the land figure F, b1.corresponding to the circular aperture figure FAC shown in FIGS.
The same applies to F and b2, and also as shown in Fig. 17 (c-
1), hexagonal aperture figure FA shown in (c-2)
The same applies to the land figures F and C1°FLc2 corresponding to O.

Therefore, the peripheral lengths of these two land figures F and a1 F are transferred to the aperture figure F, FAC1a
2 I: Select an aperture figure to be pattern matched by comparing it with the perimeter of AS.

FIG. 18 is a conceptual diagram illustrating a method of selecting an averted figure by comparing peripheral lengths. In the figure, the horizontal axis represents the peripheral length, and the three aperture figures FF and F shown in FIG.
80 AC” AS, 'AO9' AC are shown. Also, the 18th
In the figure, the peripheral length of a land figure F having a linear interpolation vector B, a+, and the peripheral length of a land figure F having a circular interpolation vector Ba1°, +. is shown a2. In comparing the peripheral lengths, first, the maximum peripheral length L and the minimum peripheral length ax Lmin, which are expressed by the following equations, are calculated.

L - Lc t ε1 (14a) aX L, = L, - ε1 (14b) an ε1: constant Here, the constant 81 is a value that takes into account the figure reading error. In the example shown in FIG. 18, the maximum peripheral length L of the land shape
The perimeter 'AS of the aX square aperture figure F exists between the minimum perimeter L1o and the minimum perimeter L1o. Therefore, the square aperture figure FAS is selected as the aperture figure to be pattern matched. Note that, generally, a plurality of 7-batch I figures are selected by comparing the peripheral lengths.

Next, the X area width and the X area width of the land figure and the aperture figure are compared, and a smaller number of aperture figures are selected from among the aperture figures selected by the above process.

As shown in FIGS. 17(a-1) and (a-2), the land figure E1. al
, the X region width ΔX and the y region width Δy are determined. Also, circular interpolation vector B. The X area width ΔXC and the y area width ΔVC are also determined for the land figure F having a.

Next, by the following formula, maximum X area width ΔX1.18x, minimum xgA area width Δxwin, and maximum X area width Δy□ax'
Minimum y region width Δy,. seek.

ΔX −Δxc+ε2 ... (15a) aX Δx, = ΔX[-ε2・(15b) 1n Δy = Δyc+ε2 ... (16a)
max Δyllio−ΔyL−ε2 (16
b) ε2: Constant The constant ε takes into account the error, similar to the constant 81 described above. FIG. 19 is a conceptual diagram showing a method of selecting an aperture figure by comparing the X area width and the X area width. As shown in the figure, maximum X area width ΔX, minimum X area width Δ
x1o, maximum y area n+ax width Δy, minimum X area width Δywin range ma
× [If the aperture t/figure has an X area width Δx and a y area width ΔyA1 within lfl, that aperture figure is selected. Note that, in general, a plurality of aperture figures are also selected by this selection.

Note that if no aperture shape is selected as a result of the above-mentioned comparison of the peripheral length and the comparison of the xgA area width and the X area width, the outline side vector data of the shape is It is transmitted from the recognition means 62 to the thinning processing means 63, and undergoes the above-mentioned thinning processing and wiring information detection processing again.

On the other hand, if one or more AB-111 figures are selected for the figure, pattern matching is performed as follows. FIG. 20 is a conceptual diagram showing a method of pattern matching between a land figure and an aperture figure.

Fig. 20 (a-1) shows the aperture figure FA represented by contour force vector 1 health data, and Fig. 20 (a-2) shows the dot data DAx obtained by converting the aperture figure FA into binary data. show. Dot data DAx.

The value of is 1'' for pixels in the black area, and ``O1 for pixels in the white area.
It is 1. Similarly, Fig. 20 (b-1) shows the land figure F represented by the contour side vector, and Fig. 20 (b-2)
) indicates the data DLx obtained by converting the land figure F into binary data. However, the land figure in FIG. 20 (b-1) is the linear interpolation vector B and the circular interpolation vector B mentioned above. I have both +0 and +. Moreover, as shown in FIG. 20 (b-2), in the land figure of the part existing on the right side with respect to the direction of movement of the linear interpolation vector B, +, the dot data DLx is "
It is 1 pa. On the other hand, linear interpolation solid 1 hell B, on the left side of +, and circular interpolation vector B. In the land figure on the right side of +, dot data 01. is indicated by an “X”. A portion where the value of the dot data DLx is "X" is a don't care region RD, and matching with the aperture figure FA is not performed in this portion. In Figures 20 (a-2) and (b-2), the size of each dot is exaggerated for ease of understanding.

In pattern matching, the following matching value ρMx and non-matching value PN are used for each pixel Px.
Squeeze xy and J1.

21 when DAx,=1 and D L xy=1
M=1. PN=O-(17a)xy
xy DA xy-1 and DL xy= O's edge or when DAxy=0 and DLxy==1: PMx,
=0. PNx,=1-(17b) Also, the following matching ratio MR is calculated from these values.

Σ ΣP M xy −L If the land figure F is similar to the aperture figure FA, the matching ratio MR will be small. Therefore, if the matching ratio MR is smaller than a predetermined threshold value, it is determined that the land figure FL matches its aperture figure "." In this pattern matching, as shown in FIG. , (b-2), in the small don't care area RD, the RI calculations of equations (17a), (17b), and (18) are not performed. Therefore, the correct aperture is also obtained for the deformed land figure F from which the wiring figure has been removed. Figures can be found and highly reliable pattern matching is possible.

Incidentally, if the matching ratio MR is equal to or higher than a predetermined threshold as a result of performing pattern matching on one aperture shape, pattern matching is performed on other preselected aperture shapes. If it is determined that the land figure matches any aperture figure, land figure CAD data D1 including the aperture number indicating the aperture figure and the center coordinates of the land figure is generated, and the aperture recognition means 62 generates the land figure CAD data D1. transmitted to station 1.

On the other hand, when the figure represented by the contour side vector data CBD given to the aperture recognition means 62 does not match any aperture figure, the contour side vector data CBD is transmitted to the thinning processing means 63. and,
The line thinning process and wiring information detection process described above will be performed again.

Note that since the land figure F is obtained by reading the artwork figure by the imager 11, the shape of its boundary portion is often quite different from that of the aperture figure. Therefore, in calculating the above-mentioned matching ratio MR, it is also possible to adopt a method of lowering the weight of the boundary of the land figure. FIG. 21(a) shows dot data DLx of land figure F.

The value of one pixel layer at the outermost periphery of the land figure F[ is "1", and the value of the inner pixel is "2". FIG. 21(b) shows dot data DA of the aperture figure FA corresponding to this.

shows. In the case shown in Figure 21, (17a), (
The matching value ρMx given by equation 17b) and the non-matching value PN are given as shown in Table 1 below.

Table 1 Dot data DAx as shown in FIG.

If you perform pattern matching using DL',
Since the matching value PM at the center of the land figure F becomes large, the reliability of pattern matching can be improved even when the shape of the boundary of the land figure F is quite different from the aperture figure FA. In addition, Dora 1-
The pixels whose data is ``1'' may be placed in two layers instead of one layer on the outermost periphery as shown in FIG.
If the value of the dot data in the pixel of the layer is set to "'0.1" and the value of the dot data in the pixel inside the layer is set to 1", pattern matching can be performed with further weight placed on the central portion.

C9 implementation procedure Figure 22 shows the CAD process that combines each process explained above.
This is a flow rate showing the implementation procedure for creating data. Also, FIG. 23 shows an example of changes in artwork figures processed according to this procedure. i is a conceptual diagram.

First, in step S1, avergent figure data for land, wiring width data, maximum aperture area FA
, and the minimum aperture area FAmanlax, etc., are input by the operator from the external input means 2. In this example, as shown in FIG.
1 and a circular aperture figure FA2 are input, and a wiring width T1° and a wiring width T2 are input as wiring width data. Furthermore, each aperture figure ``'' is given an aperture figure number A1, A2, respectively.

Note that the maximum aperture area FA and the minimum aperture max -dit? ia area FA. may be automatically determined by the workstation 181 A2 based on the input aperture figures F 1 and F 2 . At this time, the operator selects these aperture shapes FA,
There is no need to input FA, , n.

Next, in step S2, the artwork figure is read by the imaging means 11. FIG. 23(b) shows an example of the read figure F1. In the figure, figures F are square land figures F, 1, circular land figures F, 2, . and a wiring diagram FIJI that connects these to each other.

The image data of the figure F1 is converted into dot data D, which is 2fri data, by the scanner section 12 at step S3 and transmitted to the workstation 1.

In step S4, the dot data D is converted into contour edge vector data C0D1 at the workstation 1. The contour blade vector CB represented by the converted contour blade vector data CBD1 is shown in FIG. 23(c).

Next, in step S5, the isolated point area detection means 61 compares the figure F1 with the minimum aperture area FA· inputted in step S1.

If the l11n figure is smaller than the minimum aberration 17 area FAwin, the process moves from step S5 to step S6, and error information E is transferred from the isolated point area detection means 61 to the workstation 1 and further displayed on the CRT 3.

When the operator who has viewed the error information E displayed on the CRT 3 determines that the figure is unnecessary, he erases the outline edge vector data of the figure in step S7. on the other hand,
If the figure is required, an instruction is given to treat it as a normal figure with necessary modifications. In this case, the process returns to the procedure after step S5.

The figure F1 in FIG. 23(b) has the minimum aperture jz fl
Since it is larger than the J area FA1n, the process moves from step S5 to step S8, where it is compared with the maximum aperture area FA. If the figure to be compared is smaller than the maximum All1a x percher area FAIIlax, the process moves from Step S8 to Step 814, and this process will be described later. On the other hand, in the case of a figure larger than the maximum aperture area FAlax, such as the figure F1, the process moves from step S8 to step S9.

In step S9, the line thinning processing means 63 checks the history of each line thinning process.

This is N wiring widths T1 to TN (in the example of Fig. 23, N
=2), line thinning may be performed up to N times based on the line widths T1 to T8. At this time, the wiring widths used for thinning are selected in order from the smallest value. If thinning processing has been performed on the largest wiring width TH before reaching step S9,
The process ends in step S9, and the contour side vector data C[3D1 of the figure is transmitted as is from the thinning processing means 63 to the workstation 1 as non-land figure CAD data DR.

On the other hand, since the figure F1 in FIG. 23(c) has not been thinned even once, the thinning process is performed in step S10. In step S1°, line thinning processing is performed based on the smallest interconnect width T1. FIG. 23(d) shows the figure F* after the thinning process and its contour side vector CB''. Note that if the thinning process has been performed several times before step 810, It goes without saying that the smallest wiring width T is used among the remaining wiring widths T, ~TN that are not used for the thinning process. Then, when step 510t-thinning is performed, The thinning history of the figure F (F'') is stored in the m-line processing means 63.

Next, the process moves from step S10 to step 311, and the contour side vector data CBD of the thinned figure F19,
" is transmitted from the thinning processing means 63 to the wiring information detection means 64, and the wiring information detection processing is performed.

In step 811, it is determined that the figure F18 in FIG. 23(d) has the wiring figure FW1*, and in step 812, the wiring information is extracted. In this way, the separated wiring figure F141'' and land figure F, 1*F12* are connected to the 23rd
Shown in Figure (c). This wiring information is based on the wiring diagram F141″
The two end points of the center line P1. P[ and the refraction point P1 between
A wiring CAD consisting of coordinate data consisting of the position coordinates of , and the wiring width T1 used for thinning in step 810.
The data is D. This layout [11 CAD data D, 1 is transmitted from the wiring information detection means 64 to the workstation 1.

For shapes that are determined to have no wiring information in step 811,
In step 813, the line is subjected to inverse thinning processing. 23rd
The land figures F11" and F121 in FIG. 23(e) are subjected to IIwIA inverse processing in step S13. In FIG.
, F is shown. As shown in the figure 1[2, the contour side pect[2] representing the land figures F and F is
1L2 Data CB, 1. In CB, 2, linear interpolation vectors S1. +.

CBL2+ is supplemented.

These land figures F 1 and F 2 are compared again in step 58 with the maximum aperture area FΔ1llax. This time, since the wiring figure F, 41 is removed, the land figures FL1 and FL2 are smaller than the maximum aperture 1ν area FA, ax. Therefore, the process moves from step S8 to step 814, and aperture recognition means is performed.

In step 814, the aperture recognition means 62 selects an aperture figure to be pattern matched by selection based on the peripheral length of the figure and selection based on the X area width and the y area width, as described above. If no aperture figure is selected, the process returns from step S15 to step S9, and undergoes thinning processing and wiring information detection processing.

In the example of FIG. 23, for the land figure F[, the aperture shown in FIG. 23(a)! - The figure FA1 is selected, and for the land figure F[2, the 7-pa, cha figure FA2 is selected.

Then, in step 816, the land figure FL1.
Pattern matching is performed between FL2, aperture figure FAI, and FA2. As a result, the land figure "11" is recognized as matching the aperture figure "4."
In addition, the land figures F and 2 are recognized as matching the aperture figure A2.This pattern matching process is performed using the ton 1 to care area defined by the circular interpolation vector as described above. .Each land figure F
Aperture figures F and F that match 11 L2 and F are respectively recognized and 1
A2 Then, for each land figure F , F , the position [
1 [2 Land figure CAD data D consisting of location coordinate data and aperture figure number, 1. D,2 is generated. This land figure CAD data DL1. DL2 is transmitted from the aperture recognition means 62 to the workstation 1.

As a result of the above processing, for the figure F1 in FIG.
1. L2 Land figure CAD data D, 1. D,2 are generated. These CAD data, along with CAD data of other figures, are output from the work stage 1 to an output means 5 such as a magnetic tape as necessary.

FIGS. 24 to 28 are conceptual diagrams showing processing procedures for figures "2" to "6" that are different from FIG. 23.

Note that the processing procedure for these figures F2 to F6 is the same as that in Figure 22, and the aperture figure and wiring width data are the same as those in Figure 22.
The one shown in Figure 3(a) is used.

FIG. 271(a) shows the figure F read in step S2 of FIG. Figure ``2 is Stella 183
．． 84, the contour side vector CB2 is obtained as shown in FIG. 24(b). Also, figure F is shown in Figure 23(a)
matches the aperture figure FA1, so step 85
．． 38 in red and white, and step S14. S15 and 816
It is recognized that the area coincides with the avert 11 area FA1.

Figure 25(a) +7) Figure F3 is 2311g(b)
),! :Inji rad figure F, 1. F, 2, and a wiring figure FW2 having a wiring width T2 that connects them in a H shape.

As shown in FIG. 25(b), the figure ``3'' is converted into a contour side vector CB3 in step S4.Furthermore, after steps 88 and 89, thinning processing is performed in step 310 using the wiring width T1. As a result, the thinned figure F3 shown in FIG. 125(c) is obtained.

However, since the wiring figure F, 42 of the figure F2 has a wiring width T2 larger than the wiring width T1 used for this thinning, the wiring information of the thinning figure F3'' is not detected in step 811. As shown in FIG. 25(d), the thinning inverse process is performed in step S13, resulting in the original figure F3.In this way, for a figure from which the wiring figure is not extracted, the thinning figure F3'' is actually It is assumed that substantial line thinning inverse processing has been performed by using the contour side vector CB3 created in step $4 before the a-line conversion without performing line thinning inverse processing. After that, step S10 is performed again via steps 88 and 89.
In this step, the actual 2'rm wiring is performed using the wiring width T2. This thinned figure “3” is the second
This is shown in Figure 5(c). In step S11, the wiring figure FW2 is detected from the thin line -2*2 figure "." As shown in FIG.

The land figure FL1$2 FL2 is then subjected to thinning inverse processing in step S13 (
With Figure 25 (g)), then step 38゜81
4. At S15.816, it is recognized that they match the aperture figures F and F, respectively, and this A1 A
2. CAD data of these land shapes is obtained.

FIG. 26(a) shows the maximum aperture t・region 1! ! 1 FΔ
A figure F4 larger than max is shown. Regarding this figure "4," in step S4, the contour side vector CB4
(FIG. 26(b)), and after passing through step 38° $9, in step 810, the wiring width T1 is first determined.
Thinning processing is performed using (FIG. 26(c)).

However, since no wiring information is detected, the process proceeds from steps 811 to 813, where the thinning inverse process is performed and the figure F4 (FIG. 26(d)) is returned to its original size. Further step 8
8.89, in step S10, thinning processing is performed using the wiring width T2 (see FIG. 26 (0).
)). However, the wiring information is still not detected, so
In step S13 again, the process returns to the original figure F4 (
Figure 26(f)). Next, when the process reaches step S9 via step S8, it becomes clear from the line thinning history that the line thinning process has been performed for all of the previously input wiring widths T, T, . Therefore, in step S9, the figure F
4 is completed, and the contour edge vector data CB4 is obtained as is as non-land figure CAD data DIl.

The figure F5 in FIG. 27(a) is composed of a land figure F12 that coincides with the aperture figure F1, and a wiring figure F15 drawn as a circle with a wiring width T2 surrounding it. In step S4, the contour edge vectors 1 to CB, 2 are obtained for the land figure FL2, and the wiring figures F, 1
5, clockwise closed-loop contour blade vector data CB141 and counterclockwise closed-loop contour blade vector CB141 and counterclockwise closed-loop contour blade vector CB2 are obtained inside the clockwise closed-loop contour blade vector data CB141. Contour side vector f-ta CBD for figure F5, 11. COD, 42. C.B.
The connected state of D12 is shown in FIG. 27(b), and the closed loop vector data CB D C[ corresponding to the contour blade vector data C13, 42, 0B 12, respectively.
3D, 2 are connected to the contour blade vector data CBD, 1 corresponding to the outermost W2/contour blade vector CB, 41, respectively. After steps 88 and 89, this figure F5 has a wiring width T in step S10.
(FIG. 27(c)). As a result, the thinned figures F, l! ]* is step 3
11, it is detected as a wiring diagram. Then, in step 812, the wiring information (wiring conversion data) is extracted, and the wiring figure F, 4C, * and the land figure F, 2* are separated (FIG. 27(d)). At this time, as shown in FIG. 27(d), contour blade vector data CBD, CBD'' representing the wiring figure FW5*, and closed loop vector data CB representing the land figure FL29.
D, 2* are separated. Then, in step 813, only the land figure F,2* (c8D,2') is subjected to thinning inverse processing to become the original land figure F,2 (second
Figure 7(e)). This land figure F[2 is
At steps 314, 815, and 816, it is recognized that it matches the aperture figure F^2, and land figure CAD data is obtained.

FIG. 28(a) shows a figure F6 drawn with a wiring width T2. The figure F6 has a shape that is a combination of a circle and a cross. Regarding this figure F6, five contour edge vector data CBD11 to CBD15 are obtained in step S4 (FIG. 28(b)). The clockwise contour blade vector CB11 represents the outline of the black area of the figure F6, and the other counterclockwise contour blade vectors CB12 to CB is represent the outline of the white area inside thereof. The connection state of the leap loop vector data CBD11 to CBD15 regarding these contour blade vectors CB11 to CB15 is
This is shown in Figure 8(b). However, regarding this figure F6, closed loop vector data CBD11 to C as shown in FIG.
Separation of BD15 is not performed.

The figure F6 in FIG. 28(b) is drawn in step S8. After S9, in step 810, thinning is performed using the wiring width T1. However, since the original figure F6 is drawn with the wiring width T2, the thinned figure F6 in Figure 28 Book 1 (c) is drawn in step S11.
Wiring information is not detected.

Umbrella 1 - Therefore, the thinned figure FG is subjected to inverse thinning processing in step S13 (FIG. 28(d)). However, in this case,
As described above, the original contour blade vector CB - CB1. created in step S4. is simply used again.

After steps 88 and 89, in step S10, m5it processing using the wiring width T2 is performed (28th
Figure (0)), the resulting thinned figure F682 is
In step 811, all of the information is detected as wiring information, and in step 812, wiring CAD data is generated.

As described above, in this embodiment, when the artwork figure read by the image sensor r11 is composed of a layout figure and a land figure, one of the figures is separated and the wiring ( AD data, land shape CA
D7''-Yu is generated. Also, for figures that are not wiring figures or land figures, their contour side vector data is directly qqed as non-mounted figure CAD data. Also, there is an advantage that CAD data composed of wiring CAD data, land figure CAD data, and non-land figure CAD data can be easily obtained.

D. Modification ■ In the above two embodiments, the line thinning process was performed by reducing the outline side of the figure by 1/2 of the predetermined wiring width over the entire circumference. Not limited to
For example, the wiring width may be reduced by 1/4. However, if the wiring width is reduced by 1/2 of the predetermined wiring width as in this embodiment, there is an advantage that the wiring information can be easily reduced to <rl by one thinning.

Also, instead of thinning processing using contour edge vectors,
For example, by performing line thinning processing using the dot l~ data of the figure and a so-called line thinning mask (for example, a 3x3 mask), the width of the wiring figure part is reduced to one pixel, and its center line is It's just as good as you ask.

(2) In the above embodiment, the artwork AW is read by the dancing image means 11 and CAD data is obtained based on the image data, but the reading of the image data and the subsequent processing are performed separately. In other words, CAD data can be obtained by processing image data that has been read and stored in advance.

[Effects of the Invention] As described above, according to the present invention, aperture shape data indicating the shape of a land figure and wiring width data indicating the line width of a wiring figure are registered in advance, and I/O of an artwork figure is registered in advance.
After converting the artwork shape to a wiring diagram,
In addition to separating land figures and other figures, CAD data is generated by separating wiring CAD data, land figure CAD data, and non-land figure CAD data, making it easy to create CAD data for artwork figures. There is an effect that can be obtained.

[Brief explanation of the drawing]

Fig. 1 is a schematic configuration diagram of a bagging system that creates CAD data by applying the sharp tMPA of the present invention, Fig. 2 is a 1R conceptual diagram showing a method for generating contour side vectors, and Fig. 3 is the fourth flow y-p-1-1 showing the procedure.
The figure is a conceptual diagram showing the mask used to generate the contour side vector. Figure 5 is a conceptual diagram showing the method of creating the contour side vector using the polygonal line approximation method. Figure 6 is the configuration of the idle loop vector data. FIGS. 7 and 8 are conceptual diagrams showing contour side vectors and their connection states. FIG. 9 is a conceptual diagram showing the maximum aperture t/area. FIG. 10 is a method for separating contour side vectors. Figure 11 is a conceptual diagram showing the procedure for detecting wiring information g8, Figures 12 to 15 are conceptual diagrams showing the ability and method of detecting wiring information by thinning, and Figure 16 is , a conceptual diagram showing a land shape after removing the wiring diagram, FIGS. 17 to 21 are conceptual diagrams showing a method of aperture 11m recognition, and FIG. Flowchart showing the procedure, Figures 23 to 2
FIG. 8 is a conceptual diagram showing changes in figures in each processing procedure. DESCRIPTION OF SYMBOLS 1... Workstation, 2... External input] means, 6... CAD data conversion means, AP... Artwork figure input device, ΔW... Artwork, F... Figure, FA・・・Averagep
Shape, ``S...Land figure, F4...Wiring figure, FR...Filled figure, CB...Contour side vector, CBD...Contour side vector data, D...Land figure CAD data , D... Wiring CAD data,

Claims (1)

[Claims] (1) A method for creating CAD data of an artwork figure having a plurality of land figures of a predetermined shape and a wiring figure of a predetermined line width that connects these land figures, the method comprising: (a) the shape of the land figure; Aperture shape data indicating the line width of the wiring figure and line width data indicating the line width of the wiring figure are registered in advance, and (b) the contour side of the artwork figure is determined based on image data obtained by reading the artwork figure. After creating the vector data, (c) thinning the contour side vector data so as to reduce the contour side over the entire circumference of the contour side of the artwork figure; (d) after this thinning. In the contour side vector data,
By extracting a vector portion having a width smaller than a predetermined reference width as a wiring vector, wiring CAD data having position data and the line width data of the wiring figure is obtained, and from the outline side vector data. Obtain non-wiring vector data from which the wiring vector is removed, and (c) pattern matching the non-wiring vector data and the aperture shape data to convert the non-wiring vector data into aperture shape data and non-aperture shape data. (f) obtaining land figure CAD data having the position data of the land figure and the aperture shape data, and obtaining non-land figure CAD data from contour side vector data of the non-aperture shape data; How to create CAD data for artwork shapes.