JP2007322354A - Silk inspection apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable a silk printing pattern which is printed on a board and be inspected with high accuracy. <P>SOLUTION: A silk inspection apparatus is equipped with an image capture section for capturing an image of the board; a silk region cut-out section for cutting out a silk printing pattern image from the image of the board acquired by the image capture section; a cut-out image storing section for storing the silk printing pattern image cut out by the silk region cut-out section; a CAD image data reading section for reading a CAD image of the silk printing pattern; a means for totaling the matching levels, between the silk printing pattern image and the CAD image read by the CAD image data reading section, while shifting them; a verification result storing section for storing the totaled result of the matching levels, in association with shifted coordinates; a position shift determining section for determining the amount of shift of the silk printing pattern, based on the totaled result stored in the verification result storing section; and a defect-determining section for determining the defects of the silk printing pattern, based on the amount of shift of the silk printing pattern that is determined by the position shift determining section. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a silk inspection apparatus.

  Conventionally, a wiring pattern made of a conductive metal such as copper is formed on a surface of a substrate such as a printed wiring board by a method such as printing plating, and a solder resist for insulation protection is further formed thereon, and further, a symbol or circuit symbol of an electronic component. A silk print pattern or the like indicating characters or lines such as is overprinted. With the miniaturization of electronic components, wiring and printing are also miniaturized. Therefore, it is necessary to inspect by some method after manufacturing the product whether or not these patterns are accurately formed and printed.

  In recent years, visual inspection of substrates has shifted from visual inspection to inspection by an inspection apparatus using an image processing apparatus in order to increase the accuracy of inspection and shorten inspection time, and perform visual inspection of a large number of substrates in a short time. It has become possible to However, the inspection performance of conventional inspection equipment is not sufficiently high, and false recognition, that is, the generation of false information is not completely eliminated. The reality is that it takes time to develop a very difficult development (see, for example, Patent Documents 1 to 3).

  In a general substrate manufacturing process, solder resists, pads, through-holes, and the like are manufactured according to a standard, so that manufacturing errors are extremely small. On the other hand, since the standard accuracy in silk printing is not so high, the deviation of the printing pattern becomes large for each substrate, and a relatively good substrate, that is, a golden board, can be used as inspection master data. Have difficulty. Therefore, the inspection of the silk print pattern cannot be performed with high accuracy, and the inspection accuracy is low.

In addition, a technique has been proposed in which a silk print pattern is extracted by color, and inspection is performed with a low inspection accuracy with respect to deviation of the silk print pattern.
JP 2002-158413 A JP 2003-86919 A JP-A-11-337498

  However, the conventional inspection apparatus cannot inspect the silk print pattern with high accuracy, and the inspection accuracy is lowered.

  The present invention solves the above-mentioned conventional problems, and based on the amount of deviation between the silk printed pattern image printed on the substrate and the CAD (Computer Aided Design) image of the silk printed pattern, the silk printed on the substrate An object of the present invention is to provide a silk inspection apparatus capable of inspecting a silk print pattern printed on a substrate with high accuracy so as to determine whether the print pattern is good or bad.

  Therefore, in the silk inspection apparatus of the present invention, an image acquisition unit that acquires an image of a substrate, a silk region cutout unit that cuts out an image of a silk print pattern from the image of the substrate acquired by the image acquisition unit, and the silk region A cutout image storage unit that stores an image of a silk print pattern cut out by the cutout unit, a CAD image data read unit that reads a CAD image of the silk print pattern, a CAD image read by the CAD image data read unit, and the silk A means for counting the degree of coincidence while shifting the image of the print pattern, a collation result storage unit for storing the tabulation result of the degree of coincidence with respect to the shifted coordinates, and silk printing from the tabulation result stored in the collation result storage unit A misregistration determination unit that determines a pattern misregistration amount, and a silk print pattern determined by the misregistration determination unit And a defect determining unit for determining a failure of the silk print pattern on the basis of the shift amount of emissions.

  In another silk inspection apparatus of the present invention, a connected component extracting unit that extracts a connected component of the CAD image, a group creating unit that divides the connected component extracted by the connected component extracting unit into a plurality of groups, and a group There is a coincidence rate totaling unit that calculates the degree of coincidence between the silk print pattern image and the CAD image by shifting each time.

  In still another silk inspection apparatus of the present invention, further, a binarization unit that binarizes the image of the silk print pattern, a vertical pixel histogram creation unit that creates a vertical pixel histogram of the binary image, A horizontal pixel histogram generation unit that generates a horizontal pixel histogram of the binary image, a CAD vertical pixel histogram generation unit that generates a vertical pixel histogram of the CAD image, and a horizontal pixel histogram of the CAD image A CAD horizontal pixel histogram creation unit, and a matching unit that calculates the degree of matching while shifting pixel histograms created from the silk print pattern image and the CAD image.

  In still another silk inspection apparatus of the present invention, a connected component extracting unit that extracts a connected component of the CAD image, a group creating unit that divides the connected component extracted by the connected component extracting unit into a plurality of groups, A CAD vertical pixel histogram creation unit that creates a vertical pixel histogram for each group of CAD images, a CAD horizontal pixel histogram creation unit that creates a horizontal pixel histogram for each group of CAD images, and the silk print pattern A binarization unit that binarizes the image of the image, a vertical pixel histogram generation unit that generates a vertical pixel histogram of the binary image of the silk print pattern, and a horizontal pixel histogram that generates a horizontal pixel histogram of the binary image Pixels created from the direction pixel histogram creation unit and each of the silk print pattern image and CAD image And a verification unit for calculating a coincidence degree by shifting the Sutoguramu each other.

  In still another silk inspection apparatus of the present invention, the coincidence rate counting unit further includes a binary image obtained by binarizing the image of the silk print pattern and a binary image of the CAD image. Among the pixels that are common to each other, the coincidence rate is totaled as a pixel in which a pixel that is black or a pixel that is white is the same.

  In still another silk inspection apparatus of the present invention, the coincidence rate counting unit further compares a difference in feature values between pixels of the silk print pattern image and the CAD image with a threshold value, Count whether they match.

  According to the present invention, the silk inspection apparatus determines whether the silk printing pattern printed on the substrate is good or bad based on the amount of deviation between the silk printing pattern image printed on the substrate and the CAD image of the silk printing pattern. It is like that. Thereby, the silk print pattern printed on the substrate can be inspected with high accuracy.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing the configuration of the silk inspection apparatus according to the first embodiment of the present invention.

  In the figure, reference numeral 11 denotes a substrate such as a printed circuit board to be inspected. The substrate 11 is made of resin, ceramics, or the like, and a wiring pattern (not shown) made of a conductive metal such as copper is formed on one side or both sides thereof. Various components are connected and mounted on one or both surfaces of the substrate 11 by soldering. Here, it is assumed that the substrate 11 is in a state before the components are mounted. The component is, for example, an electric component or an electronic component such as a capacitor, a resistor, a filter, a semiconductor, or an IC (Integrated Circuit), but may be any type of component. Also, a solder resist for insulation protection is applied to one side or both sides of the substrate 11, and a silk print pattern indicating characters and lines such as component symbols and circuit symbols is printed.

  An image acquisition unit 12 includes an imaging device including an imaging device such as a CCD (Charge Coupled Device) and CMOS (Complementary Metal Oxide Semiconductor), and an optical device such as a lens. Get an image. In the present embodiment, an RGB (Red Green Blue) color image is acquired.

  A silk area cutout unit 13 cuts out a silk print pattern from the image of the substrate 11 acquired by the image acquisition unit 12 using the feature data stored in the cutout feature storage unit 29. Reference numeral 14 denotes a cutout image storage unit that stores an image of a silk print pattern cut out by the silk region cutout unit 13.

  A CAD database 20 stores CAD data such as a solder resist pattern, a silk printing pattern, a pattern of each copper layer, and the outer shape of the substrate 11. Reference numeral 21 denotes a silk CAD data acquisition unit which acquires CAD data of a silk print pattern from the CAD database 20. A CAD image data converter 22 converts the CAD data of the silk print pattern acquired by the silk CAD data acquisition unit 21 into a CAD image, that is, silk CAD image data.

  A CAD image data storage unit 23 stores the CAD image of the silk print pattern converted by the CAD image data conversion unit 22, that is, the data of the silk CAD image. Reference numeral 24 denotes a connected component extraction unit, which acquires a block of silk print patterns, that is, connected components, from the silk CAD image data converted by the CAD image data conversion unit 22, and labels each acquired connected component with a label ( Number).

  Further, 15 is an alignment unit, and the silk print pattern image stored in the cut-out image storage unit 14 and the print pattern according to the shift range of the data of the silk CAD image stored in the collation coordinate storage unit 15a in advance. Align with the silk CAD image.

  In this embodiment, the shift range is 3 on the X axis and 3 on the Y axis, that is, the shift range is 3 × 3. Then, with the 3 × 3 upper left (X = 1, Y = 1) as the starting point, the position is shifted to nine locations up to the lower right (X = 3, Y = 3) while moving in the X direction. Further, as the shifted coordinate numbers, the upper left (X = 1, Y = 1) is “1”, and the lower right (X = 3, Y = 3) is “9”.

  Reference numeral 16 denotes a collation unit that superimposes the silk print pattern image stored in the cut-out image storage unit 14 and the silk CAD image of the print pattern in accordance with the coordinates of the alignment unit 15. Reference numeral 17 denotes a coincidence rate counting unit, which calculates and tabulates the degree of overlapping of the silk print patterns, that is, the degree of overlapping, in the image superimposed by the matching unit 16. For example, the degree of overlap can be represented by a number obtained by dividing the number of pixels of the overlapping portion of the silk print pattern by the total number of pixels of the image.

  Reference numeral 18 denotes a collation result storage unit, which stores a result obtained by collating the result totaled by the coincidence rate totaling unit 17 with the array of the shifted coordinate number P1. Reference numeral 19 denotes a collation coordinate changing unit which changes the shifted coordinate from the upper left (X = 1, Y = 1) to a point (X = 2, Y = 1) moved in the X direction.

  Then, the positioning unit 15, the matching unit 16, the matching rate totaling unit 17, the matching result storage unit 18, and the matching coordinate changing unit 19 repeat the above-described processing until the nine positions that are the shift ranges are completed, After the process for the nine places is completed, the process from the misregistration determination unit 25, which is the subsequent process, is performed.

  Here, the misregistration determination unit 25 selects the portion having the highest degree of coincidence among the nine shifted coordinate results stored in the matching result storage unit 18 as the silk print pattern image and the silk print pattern. It is determined that the amount of deviation from the CAD image.

  Reference numeral 26 denotes a defect determination unit, which compares the degree of coincidence of the shift portion determined by the position shift determination unit 25 with a threshold value to determine whether or not there is a defect in the silk print pattern. Reference numeral 27 denotes a determination result storage unit that stores the determination result of the defect determination unit 26. A result output unit 28 outputs the determination result stored in the determination result storage unit 27 to a file or displays it on the screen.

  Next, the operation of the silk inspection apparatus having the above configuration will be described.

  FIG. 2 is a diagram showing an image of a silk print pattern and a CAD image in the first embodiment of the present invention, and FIG. 3 is a diagram showing a shift range in the first embodiment of the present invention.

  First, the image acquisition unit 12 acquires a color image of the substrate 11 captured by the imaging device. The silk area cutout unit 13 cuts out the silk print pattern from the image of the substrate 11 acquired by the image acquisition unit 12 using the silk print pattern feature data stored in the cutout feature storage unit 29 in advance. A silk print pattern image 40 as shown in FIG. 2A is stored in the cut-out image storage unit 14.

  Further, the silk CAD data acquisition unit 21 acquires CAD data of the silk print pattern from the CAD database 20. Then, the CAD image data conversion unit 22 converts the CAD data of the silk printing pattern into image data, and stores the converted silk CAD image 41 data as shown in FIG. 2B in the CAD image data storage unit 23. To do. The connected component extracting unit 24 acquires a block of silk print patterns, that is, connected components, for the silk CAD image 41 stored in the CAD image data storage unit 23, and assigns a label (number) to each acquired connected component. wear.

  Subsequently, the alignment unit 15 inputs the silk print pattern image 40 and the silk CAD image 41, and stores the silk print pattern CAD image as shown in FIG. 3 stored in the collation coordinate storage unit 15a in advance. The data of the silk CAD image 41 is shifted according to the numbers “1” to “9” of the data shift range 44. In the present embodiment, it is assumed that the shift range 44 is 3 on the X axis and 3 on the Y axis, that is, within a range of 3 × 3. Then, starting from the 3 × 3 upper left “1” (X = 1, Y = 1), the 9 points from the lower right “9” (X = 3, Y = 3) while moving in the X direction Shall be shifted. Further, as the shifted coordinate numbers, the upper left (X = 1, Y = 1) is “1”, and the lower right (X = 3, Y = 3) is “9”. In the present embodiment, one block in the shift range 44 corresponds to one pixel.

  Next, the operation of the alignment unit 15 will be described using the pixel 42 of the silk print pattern image 40 and the pixel 43 of the silk CAD image 41 as an example.

  FIG. 4 is a diagram for explaining the operation of the alignment unit in the first embodiment of the present invention. FIG. 5 is a diagram for explaining the operation of the alignment unit and the collation unit in the first embodiment of the present invention. 6 is a diagram showing the collation result in the first embodiment of the present invention, and FIG. 7 is a diagram for explaining the operation of the defect determination unit in the first embodiment of the present invention.

  Here, when the pixel 42 of the silk print pattern image 40 is at the position “5” of the shift range 44, the pixel 48 shown in FIG. 4 is displayed, and the pixel 43 of the silk CAD image 41 is shifted by “ The pixel 47 shown in FIG. 4 corresponds to the position “1”. The images at this time are the silk CAD image 50 and the silk print pattern image 51 shown in FIG.

  First, the alignment unit 15 is aligned with the position shown in FIG. 5A, and the collation unit 16 is a pixel in a portion where the pixels of the silk print pattern image 51 and the pixels of the silk CAD image 50 match. The number is output to the coincidence rate counting unit 17. Then, the coincidence rate totaling unit 17 calculates the coincidence rate using the total number of pixels in the silk CAD image 41 as the denominator and using the number of matching pixels as a numerator. Then, it is stored in the collation result storage unit 18 in the array of the number “1” (X = 1, Y = 1) in the shift range 44.

  Subsequently, when the data is stored in the matching result storage unit 18, the matching coordinate changing unit 19 changes the coordinate value to the coordinate value corresponding to the position “2” in the shift range 44. An image obtained by the alignment unit 15 shifting the pixel 43 of the silk CAD image 41 and matching the coordinates of “2” in the range 44 is a silk CAD image 52 shown in FIG. Subsequently, the collation unit 16 and the coincidence rate totaling unit 17 repeat the above-described operation, and store the data in the collation result storage unit 18 in the array of the number “2” (X = 2, Y = 1) in the shift range 44. .

  Thereafter, the positioning unit 15, the collation unit 16, the matching rate totaling unit 17, the collation result storage unit 18, and the collation coordinate changing unit 19 complete the above process, and “9” (image 54) of the shift range 44 is completed. Do until. Then, after the processing for nine locations from “1” to “9” is completed, the processing from the misregistration determination unit 25 as the subsequent processing is performed. Note that the results of the shifted coordinates for nine locations from “1” to “9” are stored in the collation result storage unit 18 as shown in the table of FIG.

  Then, based on the result of the nine shift coordinates stored in the collation result storage unit 18, the position deviation determination unit 25 receives the number “5” (X = 2, Y = 2), which is the highest match rate. It is determined that the silk print pattern image 40 and the silk CAD image 41 are positions where the silk print pattern shift correction has been performed. Subsequently, the failure determination unit 26 compares the coincidence rate of the number “5” with a preset threshold value, and determines that it is not defective if the matching rate is equal to or higher than the threshold value, and determines that it is defective if it is less than the threshold value. To do. For example, in the example shown in FIG. 6, since the matching rate of the number “5” is 100 [%], the failure determination unit 26 determines that it is not defective. The result determined by the defect determination unit 26 is stored in the determination result storage unit 27. The result output unit 28 outputs the result stored in the determination result storage unit 27 as a file or displays it on the screen.

  Thus, in the present embodiment, the silk print pattern image 40 is cut out from the image of the substrate 11, and the silk print pattern shift is performed while shifting the silk CAD image 41 within a specific range with respect to the silk print pattern image 40. Corrections are made and inspections are carried out after corrections. This enables inspection by silk pattern matching, which has been difficult in the past, and realizes silk print pattern inspection with high accuracy by pattern matching, which has been difficult to inspect with high printing deviation and high accuracy. be able to.

  Next, a second embodiment of the present invention will be described. In addition, about what has the same structure as 1st Embodiment, the description is abbreviate | omitted by providing the same code | symbol. The description of the same operation and the same effect as those of the first embodiment is also omitted.

  FIG. 8 is a block diagram showing a configuration of a silk inspection apparatus according to the second embodiment of the present invention.

  The silk inspection apparatus in the present embodiment has a group creation unit 30 as shown in the figure. The group creation unit 30 divides all silk printing patterns of one substrate 11 into, for example, five groups with respect to a block of silk printing patterns acquired by the connected component extracting unit 24, that is, a label (number) for each connected component. In this case, the group numbers (in this case, any number from “1” to “5”) are attached together so that there are five groups.

  In addition, the alignment unit 15 performs image alignment for the group created by the group creation unit 30 according to the shift range for each group, as in the first embodiment. Furthermore, the collation result storage unit 18 stores the results in a two-dimensional array of group numbers and shift numbers.

  Then, the positioning unit 15, the matching unit 16, the matching rate totaling unit 17, the matching result storage unit 18, and the matching coordinate changing unit 19 perform each process until the nine shift positions are finished, After the processing is completed, processing from the misregistration determination unit 25, which is subsequent processing, is performed.

  Since the configuration of other points is the same as that of the first embodiment, description thereof is omitted.

  Next, the operation of the silk inspection apparatus in the present embodiment will be described.

  FIG. 9 is a first diagram showing an operation of superimposing images in the second embodiment of the present invention, and FIG. 10 is a second diagram showing an operation of overlaying images in the second embodiment of the present invention. FIG. 11 is a diagram showing a collation result in the second exemplary embodiment of the present invention.

  In the first embodiment, there is a case where the image shift cannot be corrected with high accuracy. For example, an image of a silk print pattern and a silk CAD image are 60 as shown in FIG. 9A and 63 as shown in FIG. 9B, respectively, over a wide range. In this case, the first patterns 61 and 64 and the second patterns 62 and 65 are distributed, respectively. In this case, when the first pattern 64 and the second pattern 65 of the silk CAD image 63 are shifted under the same conditions in order to determine the coincidence rate between the silk print pattern image 60 and the silk CAD image 63, the silk print pattern image Even if the first pattern 61 of 60 and the first pattern 64 of the silk CAD image 63 match, the second pattern 62 of the silk print pattern image 60 and the second pattern 65 of the silk CAD image 63 may not match. . That is, as shown in FIG. 9C, in the superimposed image 66, even if the first pattern 67 is matched, the second pattern 68 may be shifted. This is because the silk print pattern is not evenly displaced. In the present embodiment, a group creation unit 30 is provided in order to solve such a problem.

  Then, as shown in FIG. 10, the group creation unit 30 divides the patterns close to the two reference points 79a and 79b into two groups for the silk CAD image 73. It can be divided into two. Then, for each of the group 74 and the group 75, as in the first embodiment, the alignment unit 15, the collation unit 16, the coincidence rate totaling unit 17, the collation result storage unit 18, and the collation coordinate changing unit 19 Process. However, the collation result storage unit 18 stores nine collation results from “1” to “9” in the shift range 44 for each group 74 and 75.

  Here, it is assumed that the results of nine shift coordinates for each of the groups 74 and 75 stored in the collation result storage unit 18 are as shown in Table 80 shown in FIG. In this case, based on the result of Table 80, the position deviation determination unit 25 determines that the number “5” (X = 2, Y = 2), which is the location with the highest matching rate among the nine locations of the shifted coordinates of the group 74, is silk. It is determined that the pattern 71 of the print pattern image 70 and the group 74 of the silk CAD image 73 are positions where the deviation from the silk print pattern is corrected. In addition, the number “8” (X = 2, Y = 3), which has the highest matching rate among the nine shift coordinates of the group 75, has the pattern 71 of the silk print pattern image 70 and the group 74 of the silk CAD image 73. Are determined to be positions where the shift of the silk printing pattern is corrected. As a result, as shown by the patterns 77 and 78 in the superimposed image 76, the silk print pattern image 70 and the silk CAD image 73 can be matched.

  Then, as in the first embodiment, the defect determination unit 26 compares the coincidence rate with a threshold value and determines that it is not defective if it is greater than or equal to the threshold value, and determines that it is defective if it is less than the threshold value. For example, in the case of the groups 74 and 75 in the table 80 shown in FIG. 11, since the matching rate of the number “5” of the group 74 is 100 [%], the failure determination unit 26 determines that it is not defective. The result determined by the defect determination unit 26 is stored in the determination result storage unit 27. The result output unit 28 outputs the result stored in the determination result storage unit 27 as a file or displays it on the screen.

  As described above, in the present embodiment, the silk CAD images 73 are divided into several groups, and the shift of the silk print pattern is corrected while shifting the silk CAD images 73 within a specific range for each group. . Therefore, highly accurate displacement correction can be performed even for partial displacement of the silk print pattern, and by performing inspection after the displacement correction, it is possible to perform inspection of the silk print pattern with high accuracy.

  Next, a third embodiment of the present invention will be described. In addition, about the thing which has the same structure as 1st and 2nd embodiment, the description is abbreviate | omitted by providing the same code | symbol. Also, the description of the same operations and effects as those of the first and second embodiments is omitted.

  FIG. 12 is a block diagram showing a configuration of a silk inspection apparatus according to the third embodiment of the present invention.

  As shown in the figure, the silk inspection apparatus according to the present embodiment includes a binarization unit 31, a vertical pixel histogram creation unit 32, a horizontal pixel histogram creation unit 33, a CAD vertical pixel histogram creation unit 34, and a CAD horizontal A direction pixel histogram creation unit 35 is included. In addition, the coincidence rate totaling unit 17 and the connected component extracting unit 24 are omitted.

  The binarization unit 31 binarizes the silk print pattern image cut out by the silk region cutout unit 13. In addition, the vertical pixel histogram creation unit 32 creates a histogram by accumulating pixels in the vertical direction for the binary image of the silk print pattern stored in the cutout image storage unit 14. On the other hand, the horizontal pixel histogram creation unit 33 creates a histogram by accumulating pixels in the horizontal direction for the silk-printed pattern binary image stored in the cut-out image storage unit 14.

  Also, the CAD vertical direction pixel histogram creation unit 34 creates a histogram by accumulating pixels in the vertical direction for the data of the silk CAD image stored in the CAD image data storage unit 23. Further, the CAD horizontal direction pixel histogram creation unit 35 creates a histogram by accumulating pixels in the horizontal direction for the silk CAD image data stored in the CAD image data storage unit 23.

  The alignment unit 15 then compares the histogram created by the vertical pixel histogram creation unit 32 and the histogram created by the CAD vertical pixel histogram creation unit 34 according to the shift range stored in the collation coordinate storage unit 15a. Perform alignment. In the present embodiment, the shift range is 3 horizontally.

  Further, the alignment unit 15 determines whether the histogram created by the horizontal pixel histogram creation unit 33 and the histogram created by the CAD horizontal pixel histogram creation unit 35 are in accordance with the shift range stored in the collation coordinate storage unit 15a. Perform alignment. In the present embodiment, the shift range is 3 in the vertical direction.

  Further, the collation unit 16 collates the histogram created by the vertical pixel histogram creation unit 32 with the histogram created by the CAD vertical pixel histogram creation unit 34 according to the coordinates of the alignment unit 15. . The collation unit 16 collates the histogram created by the horizontal pixel histogram creation unit 33 with the histogram created by the CAD horizontal pixel histogram creation unit 35. In this embodiment, the collation method is represented by histogram correlation, the vertical pixel histogram correlation is the vertical histogram correlation result, and the horizontal pixel histogram correlation is the horizontal histogram correlation result. To do.

  Then, the collation coordinate changing unit 19 moves one shift coordinate in the horizontal direction with respect to the CAD vertical pixel histogram, and moves one shift coordinate in the vertical direction with respect to the CAD horizontal pixel histogram.

  The misregistration determination unit 25 also determines the coordinates having the highest correlation with respect to each of the vertical histogram correlation result and the horizontal histogram correlation result of the shift coordinates stored in the matching result storage unit 18 as a silk print pattern image. And the silk CAD image.

  Further, the defect determination unit 26 calculates the degree of coincidence between the silk print pattern image and the silk CAD image with respect to the shift portion determined by the position shift determination unit 25, and determines whether or not there is a defect in the silk print pattern. Judging.

  The determination result storage unit 27 stores the determination result of the defect determination unit 26. Then, the result output unit 28 outputs the result stored in the determination result storage unit 27 to a file or displays it on the screen.

  Since the configuration of other points is the same as that of the first embodiment, description thereof is omitted.

  Next, the operation of the silk inspection apparatus in the present embodiment will be described.

  FIG. 13 is a diagram showing a silk print pattern histogram according to the third embodiment of the present invention, FIG. 14 is a diagram showing a silk CAD image histogram according to the third embodiment of the present invention, and FIG. It is a figure which shows the correlation with the histogram of the silk printing pattern in 3rd Embodiment, and the histogram of a silk CAD image.

  First, the binarization unit 31 binarizes the silk print pattern image cut out by the silk region cutout unit 13 and stores it in the cutout image storage unit 14. Then, the vertical pixel histogram creation unit 32 creates a vertical pixel histogram 82 for the binary image 81 as shown in FIG. 13 stored in the cutout image storage unit 14. In this case, the vertical pixel histogram 82 is created by accumulating the number of binarized black pixels (portion of the binary image 81) in the direction indicated by the arrow 87, that is, in the vertical direction. Further, the horizontal pixel histogram creation unit 33 creates a horizontal pixel histogram 83 for the binary image 81 as shown in FIG. 13 stored in the cutout image storage unit 14. In this case, the horizontal pixel histogram 83 is created by accumulating the number of pixels of binarized black pixels (part of the binary image 81) in the direction indicated by the arrow 88, that is, in the horizontal direction.

  Similarly, the CAD vertical pixel histogram creation unit 34 creates a CAD vertical pixel histogram 85 for the silk CAD image 84 as shown in FIG. 14 stored in the CAD image data storage unit 23. In this case, the CAD vertical pixel histogram 85 is created by accumulating the number of binarized black pixels (part of the silk CAD image 84) in the vertical direction. In addition, the CAD horizontal pixel histogram creation unit 35 creates a CAD horizontal pixel histogram 86 for the silk CAD image 84 as shown in FIG. 14 stored in the CAD image data storage unit 23. In this case, the CAD horizontal pixel histogram 86 is created by accumulating the number of binarized black pixels (part of the silk CAD image 84) in the horizontal direction.

  Subsequently, the alignment unit 15 shifts the vertical pixel histogram 82 and the CAD vertical pixel histogram 85 with respect to the shift range of the data of the silk CAD image stored in the collation coordinate storage unit 15a in advance. Calculate In this case, as shown in FIGS. 15A to 15C, the correlation is calculated while the CAD vertical pixel histogram 85 is shifted to the positions of the lines 91, 94, and 96.

  Similarly, the correlation is calculated while shifting the horizontal pixel histogram 83 and the CAD horizontal pixel histogram 86 with respect to the shift range of the data of the silk CAD image stored in the collation coordinate storage unit 15a in advance. In this case, as shown in FIGS. 15D to 15F, the correlation is calculated while the CAD horizontal pixel histogram 86 is shifted to the positions of the lines 98, 100, and 102.

  Subsequently, the positional deviation determining unit 25 determines that FIG. 15B, which is the result having the highest correlation among FIGS. 15A to 15C, is a lateral deviation, and FIG. ) To (f), which is the result having the highest correlation, is determined as a vertical shift.

  Then, the defect determination unit 26 calculates the coincidence rate between the silk print pattern and the silk CAD image with respect to the shift determined by the position shift determination unit 25 in the same manner as in the first embodiment, and determines the determination result. The result is stored in the determination result storage unit 27. The result output unit 28 outputs the result stored in the determination result storage unit 27 as a file or displays it on the screen.

  As described above, in the present embodiment, the shift amount between the silk print pattern and the silk CAD image is detected using the histogram in the vertical direction and the histogram in the horizontal direction. Therefore, the shift amount between the silk print pattern and the silk CAD image can be detected at high speed, and the inspection of the silk print pattern can be performed with high accuracy by executing the inspection after correcting the shift.

  Next, a fourth embodiment of the present invention will be described. In addition, about the thing which has the same structure as the 1st-3rd embodiment, the description is abbreviate | omitted by providing the same code | symbol. Explanation of the same operations and effects as those of the first to third embodiments is also omitted.

  FIG. 16 is a block diagram showing a configuration of a silk inspection apparatus according to the fourth embodiment of the present invention.

  Since the silk inspection apparatus in the present embodiment has a configuration in which the first to third embodiments are combined, description of each part is omitted.

  Next, the operation of the silk inspection apparatus in the present embodiment will be described.

  FIG. 17 is a diagram showing a silk print pattern histogram according to the fourth embodiment of the present invention, and FIG. 18 is a diagram showing a silk CAD image histogram according to the fourth embodiment of the present invention.

  First, the binarization unit 31 binarizes the silk print pattern image cut out by the silk region cutout unit 13 and stores it in the cutout image storage unit 14. Then, the vertical pixel histogram creation unit 32 creates a vertical pixel histogram 110 for the binary image 105 as shown in FIG. 17 stored in the cutout image storage unit 14. In this case, the vertical pixel histogram 110 is created by accumulating the number of binarized black pixels (portion of the binary image 105) in the direction indicated by the arrow 87, that is, in the vertical direction. Further, the horizontal pixel histogram creation unit 33 creates a horizontal pixel histogram 111 for the binary image 105 as shown in FIG. 17 stored in the cutout image storage unit 14. In this case, the vertical pixel histogram 111 is created by accumulating the number of binarized black pixels (part of the binary image 105) in the direction indicated by the arrow 88, that is, in the horizontal direction.

  Further, as shown in FIG. 18, the group creating unit 30 divides patterns close to the two reference points 79a and 79b into two groups for the silk CAD image 73. It can be divided into two. Then, for each of the group 74 and the group 75, as in the first embodiment, the alignment unit 15, the collation unit 16, the coincidence rate totaling unit 17, the collation result storage unit 18, and the collation coordinate changing unit 19 Process.

  Subsequently, the CAD vertical pixel histogram creation unit 34 creates CAD vertical pixel histograms 115 and 116 for each of the group 74 and the group 75 of the silk CAD image 73 stored in the CAD image data storage unit 23. In this case, the CAD vertical pixel histograms 115 and 116 are created by accumulating the number of binarized black pixels (silk CAD image data portion) in the vertical direction. Further, the CAD horizontal pixel histogram creation unit 35 creates CAD horizontal pixel histograms 117 and 118 for each of the group 74 and the group 75 of the silk CAD image 73 stored in the CAD image data storage unit 23. In this case, CAD horizontal pixel histograms 117 and 118 are created by accumulating the number of binarized black pixels (part of the silk CAD image 73) in the horizontal direction.

  Subsequently, the alignment unit 15 shifts the vertical pixel histogram 110 and the CAD vertical pixel histogram 115 of the group 74 with respect to the shift range of the data of the silk CAD image stored in the collation coordinate storage unit 15a in advance. Calculate the correlation. Similarly, correlations are calculated for the vertical pixel histogram 110 and the group 75 CAD vertical pixel histogram 116.

  The alignment unit 15 shifts the horizontal pixel histogram 111 and the CAD horizontal pixel histogram 117 of the group 74 with respect to the shift range of the data of the silk CAD image stored in the collation coordinate storage unit 15a in advance. Calculate the correlation. Similarly, the correlation is calculated for the horizontal pixel histogram 111 and the group 75 CAD horizontal pixel histogram 118.

  The correlation calculation is the same as in the third embodiment, and a description thereof will be omitted. Since the subsequent operation is the same as that of the third embodiment, the description thereof is omitted. However, in the present embodiment, the position of the highly correlated result is determined as a shift for each of the groups 74 and 75, and the matching rate between the silk print pattern and the silk CAD image is determined for the determined shift. Calculation is performed in the same manner as in the first embodiment, and the determination result is stored in the determination result storage unit 27. Then, the result output unit 28 outputs the result stored in the determination result storage unit 27 as a file or displays it on the screen.

  Thus, in the present embodiment, the silk CAD image 73 is divided into several groups, and the vertical pixel histograms 115 and 116 and the horizontal pixel histograms 117 and 118 of the silk CAD image 73 are specified ranges for each group. The shift of the silk print pattern is corrected while shifting within. Therefore, high-speed and high-precision silk shift correction can be performed even for partial silk print pattern shifts, and high-precision silk print pattern inspection is realized by performing inspection after silk shift correction. can do.

  In the first to fourth embodiments of the present invention, the shift range is described as one pixel, but the present invention is not limited to this. Further, although description has been made using RGB as a feature for cutting out a silk print pattern, it can also be realized by using hue, saturation, and the like.

  In the third and fourth embodiments, the correlation is used to detect the degree of coincidence of the histograms. However, the difference can also be realized as a feature.

  Furthermore, in the second and fourth embodiments, the reference point of the group creating unit 30 is a corner point. However, the reference point is not limited thereto. For example, the reference point is the center or the upper left corner point of the image. The standard can be set freely.

  Furthermore, the present invention is not limited to the above-described embodiment, and various modifications can be made based on the spirit of the present invention, and they are not excluded from the scope of the present invention.

It is a block diagram which shows the structure of the silk inspection apparatus in the 1st Embodiment of this invention. It is a figure which shows the image and the CAD image of a silk printing pattern in the 1st Embodiment of this invention. It is a figure which shows the shift range in the 1st Embodiment of this invention. It is a figure explaining operation | movement of the position alignment part in the 1st Embodiment of this invention. It is a figure explaining operation | movement of the position alignment part and collation part in the 1st Embodiment of this invention. It is a figure which shows the collation result in the 1st Embodiment of this invention. It is a figure explaining operation | movement of the defect determination part in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the silk inspection apparatus in the 2nd Embodiment of this invention. It is a 1st figure which shows the operation | movement which superimposes the image in the 2nd Embodiment of this invention. It is a 2nd figure which shows the operation | movement which superimposes the image in the 2nd Embodiment of this invention. It is a figure which shows the collation result in the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the silk inspection apparatus in the 3rd Embodiment of this invention. It is a figure which shows the histogram of the silk printing pattern in the 3rd Embodiment of this invention. It is a figure which shows the histogram of the silk CAD image in the 3rd Embodiment of this invention. It is a figure which shows the correlation with the histogram of the silk printing pattern in the 3rd Embodiment of this invention, and the histogram of a silk CAD image. It is a block diagram which shows the structure of the silk inspection apparatus in the 4th Embodiment of this invention. It is a figure which shows the histogram of the silk printing pattern in the 4th Embodiment of this invention. It is a figure which shows the histogram of the silk CAD image in the 4th Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Board | substrate 12 Image acquisition part 13 Silk area cutout part 14 Cutout image storage part 16 Collation part 17 Matching rate totaling part 18 Collation result storage part 24 Connected component extraction part 25 Misalignment determination part 26 Defect determination part 30 Group preparation part 31 Binary Conversion unit 32 Vertical pixel histogram generation unit 33 Horizontal pixel histogram generation unit 34 CAD vertical pixel histogram generation unit 35 CAD horizontal pixel histogram generation units 40, 51, 60, 70 Silk print pattern images 41, 50, 52, 53 63, 73, 84 Silk CAD image 54 Image 74, 75 Group 81, 105 Binary image 82, 110 Vertical pixel histogram 83, 111 Horizontal pixel histogram 85, 115, 116 CAD vertical pixel histogram 86, 117, 118 CAD horizontal pixel histogram

Claims (6)

(A) an image acquisition unit for acquiring an image of the substrate;
(B) a silk region cutout unit that cuts out an image of a silk print pattern from the image of the substrate acquired by the image acquisition unit;
(C) a cut-out image storage unit that stores an image of a silk print pattern cut out by the silk region cut-out unit;
(D) a CAD image data reading unit for reading a CAD image of the silk print pattern;
(E) means for counting the degree of coincidence while shifting the CAD image read by the CAD image data reading unit and the image of the silk print pattern;
(F) a collation result storage unit that stores the total result of the matching degree with respect to the shifted coordinates;
(G) a displacement determination unit that determines a displacement amount of the silk print pattern from the total result stored in the matching result storage unit;
(H) A silk inspection apparatus comprising: a defect determination unit that determines a defect of a silk print pattern based on a shift amount of the silk print pattern determined by the position shift determination unit. (A) a connected component extraction unit that extracts a connected component of the CAD image;
(B) a group creating unit that divides the connected components extracted by the connected component extracting unit into a plurality of groups;
The silk inspection apparatus according to claim 1, further comprising: a coincidence rate totaling unit that calculates a degree of coincidence between the silk print pattern image and the CAD image by shifting for each group. (A) a binarization unit that binarizes the image of the silk print pattern;
(B) a vertical pixel histogram creation unit that creates a vertical pixel histogram of a binary image;
(C) a horizontal pixel histogram creation unit for creating a horizontal pixel histogram of the binary image;
(D) a CAD vertical pixel histogram creation unit for creating a vertical pixel histogram of the CAD image;
(E) a CAD horizontal pixel histogram creation unit that creates a horizontal pixel histogram of the CAD image;
(F) The silk inspection apparatus according to claim 1, further comprising: a matching unit that calculates a degree of matching while shifting pixel histograms created from each of the silk printed pattern image and the CAD image. (A) a connected component extraction unit that extracts a connected component of the CAD image;
(B) a group creating unit that divides the connected components extracted by the connected component extracting unit into a plurality of groups;
(C) a CAD vertical pixel histogram creation unit that creates a vertical pixel histogram for each group of CAD images;
(D) a CAD horizontal pixel histogram creation unit that creates a horizontal pixel histogram for each group of the CAD images;
(E) a binarization unit for binarizing the image of the silk print pattern;
(F) a vertical pixel histogram creation unit for creating a vertical pixel histogram of the binary image of the silk print pattern;
(G) a horizontal pixel histogram creation unit for creating a horizontal pixel histogram of the binary image;
The silk inspection apparatus according to claim 1, further comprising: a matching unit that calculates a degree of matching while shifting pixel histograms created from the silk printed pattern image and the CAD image. The coincidence rate totaling unit is configured such that when the binary image obtained by binarizing the image of the silk print pattern and the binary image of the CAD image are overlapped with each other, the pixels that are common to each other, The silk inspection apparatus according to claim 2, wherein the coincidence ratios are totaled as pixels that match white pixels. 3. The silk inspection apparatus according to claim 2, wherein the coincidence rate counting unit compares the difference between the feature values of the pixels of the silk print pattern image and the CAD image with a threshold value and totals whether or not they match.

Images