JP2008032532A - Defect detection device, defect detection method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent wrong detection relative to comparatively mild ink unevenness unrecognized as a defect by a person viewing a print, without reducing detection accuracy of the defect. <P>SOLUTION: The defect detection device includes the first generation part 100, a rough image generation part 101, the second generation part 102, an enlarging part 103 and a defect detection part 104. The first generation part 100 generates a differential image 132 between a reference image 130 and an object image 131. The rough image generation part 101 reduces the reference image 130 and the object image 131 by a bilinear method, and generates a reference rough image 133 and the object image 131, respectively. The second generation part 102 generates a differential rough image 135 between the reference rough image 133 and an object rough image 134. The enlarging part 103 enlarges the differential rough image 135 by the bilinear method, and generates a masking image 136 having the same size as the differential image 132. The defect detection part 104 masks the differential image 132 by the masking image 136, and simultaneously detects the defect. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technique for detecting defects in printed matter.

  Conventionally, a difference between an image obtained by capturing a printed matter (a target image to be inspected) and a reference image to be an inspection reference is obtained, and the degree of mismatch between the target image and the reference image is detected based on the obtained difference. Thus, techniques for detecting defects in the printed matter have been proposed. Such a technique is described in Patent Document 1, for example.

  On the other hand, when creating a printed matter, the operator of the printing machine generally finishes in consideration of the overall color balance. That is, it is common to make a specific color darker or lighter overall depending on the experience and preferences of the operator of the printing press.

  The human eye is sensitive to color mismatch in a relatively narrow range, and if such a phenomenon occurs, it is recognized as a defect. However, color inconsistency over a relatively wide range, such as gradual ink unevenness (light / dark), is insensitive and has a characteristic that it is difficult to recognize as a defect. In other words, the color shading normally given by the operator is not recognized as a defect by human eyes.

JP 2000-335062 A

  However, if the comparison between the target image and the reference image is simply compared for each pixel based on a predetermined threshold value, there is a problem that the shade of the color consciously given by the operator is also determined as a defect. . Originally, in printed matter, only what is recognized as a defect by humans should be detected as a defect. There is a problem that unrecognized ones are detected as defects.

  In order to solve such a problem, for example, it is possible to perform an adjustment for slowing a predetermined threshold value. However, in this case, the difference (mismatch) that should be detected as a defect is no longer detected, and the detection accuracy decreases.

  The present invention has been made in view of the above problems, and an object thereof is to prevent erroneous detection related to relatively gentle ink unevenness that is not recognized as a defect by a viewer of a printed matter without deteriorating the detection accuracy of the defect. And

  In order to solve the above problems, the invention of claim 1 is a defect detection apparatus for detecting a defect of a target image, an acquisition unit for acquiring a reference image and a target image, and a reference image acquired by the acquisition unit And a first generation means for generating a difference image based on the target image, a second generation means for generating a difference coarse image, and a difference coarse image generated by the second generation means by the first generation means. An enlargement unit that generates an image for masking by enlarging the size of the generated difference image, and a defect by masking the difference image generated by the second generation unit by an image for masking generated by the enlargement unit And a defect detection means for detecting.

  The invention according to claim 2 is the defect detection apparatus according to the invention of claim 1, wherein the reference coarse image is generated based on the reference image acquired by the acquisition unit and acquired by the acquisition unit. The image processing apparatus further includes a coarse image generation unit that generates a target coarse image based on the target image, and the second generation unit generates a differential coarse image based on the reference coarse image and the target coarse image generated by the coarse image generation unit. It is characterized by doing.

  The invention according to claim 3 is the defect detection apparatus according to claim 1, wherein the second generation unit generates a rough difference image based on the difference image generated by the first generation unit. It is characterized by doing.

  According to a fourth aspect of the present invention, there is provided a defect detection apparatus according to any one of the first to third aspects, wherein the mask image is subjected to a fattening process.

  The invention of claim 5 is a defect detection method for detecting a defect in a target image, based on an acquisition step of acquiring a reference image and a target image, and the reference image and target image acquired in the acquisition step. The first generation step for generating the difference image, the second generation step for generating the difference coarse image, and the difference coarse image generated in the second generation step are the same as the difference image generated in the first generation step. An enlargement step for generating a mask image by enlarging the size, and a defect detection step for detecting a defect by masking the difference image generated in the second generation step by the mask image generated in the enlargement step It is characterized by providing.

  The invention according to claim 6 is a computer-readable program, and the execution of the program by the computer is acquired by the acquisition means for acquiring a reference image and a target image, and the acquisition means. Based on the reference image and the target image, a first generation unit that generates a difference image, a second generation unit that generates a difference coarse image, and the difference coarse image generated by the second generation unit are generated in the first generation. Masking the difference image generated by the second generation means with an enlargement means for generating a mask image by enlarging the size of the difference image generated by the means, and a mask image generated by the enlargement means. And defect detection means for detecting defects, and functioning as a defect detection device for detecting defects in the target image.

  According to the first to sixth aspects of the present invention, a difference rough image is enlarged to the size of the generated difference image, a mask image is generated, and the difference image is masked by the generated mask image to cause a defect. By detecting this, it is possible to inspect while removing the influence of global ink unevenness. Accordingly, it is possible to prevent erroneous detection related to relatively gentle ink unevenness that is not recognized as a defect by a viewer without reducing the detection accuracy of the defect.

  According to the fourth aspect of the present invention, since the mask image is subjected to the thickening process, the influence of the global ink unevenness can be more effectively removed.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the accompanying drawings.

<1. First Embodiment>
FIG. 1 is a diagram showing a defect detection apparatus 1 according to the present invention. FIG. 2 is a diagram illustrating the configuration of the defect detection apparatus 1.

  The defect detection apparatus 1 includes a CPU 10, a storage unit 11 that stores data such as a program 120, an operation unit 15, a display unit 16, a communication unit 17 for connecting to a network 91, a drive device 18 that reads a medium 92, and an imaging unit 19. Is provided. The defect detection apparatus 1 has a function as a general computer.

  As shown in FIG. 2, the CPU 10 is connected to various components of the defect detection apparatus 1 by bus wiring. The CPU 10 operates according to the program 120 to generate control signals and control these various configurations. Further, the CPU 10 has a function of calculating various information.

  The storage unit 11 includes a read-only ROM 12 that mainly stores a program 120, a RAM 13 that is used as a temporary working area of the CPU 10, and a hard disk 14 that stores a relatively large amount of data. Various information used and generated in the defect detection apparatus 1 is appropriately stored in the storage unit 11.

  The operation unit 15 includes a keyboard and a mouse (not shown), and is operated by the operator to input information to the defect detection apparatus 1.

  The display unit 16 is a general display and displays various types of information on the screen as images. In particular, the display unit 16 displays the detection result of the defect detection apparatus 1. Thereby, the operator can confirm the detection result of the defect detection apparatus 1.

  The communication unit 17 has a function as a network interface, and connects the defect detection apparatus 1 to the network 91. By providing the communication unit 17, the defect detection device 1 can perform data communication with an external device (not shown) via the network 91. The network 91 corresponds to a public line network, the Internet, or a LAN, but is not limited to this.

  The drive device 18 is, for example, an FD drive device, a CD-ROM drive device, a DVD drive device, or the like, and has a function of reading information stored in a medium 92 that is a portable recording medium. The drive device 18 mainly transfers information read from the medium 92 to the hard disk 14.

  In particular, in the defect detection device 1 according to the present embodiment, the drive device 18 reads image data (FIG. 3: reference image 130) stored in the medium 92. That is, the drive device 18 has a function corresponding to the acquisition unit of the present invention. The reference image 130 may be acquired by the communication unit 17 receiving it from an external device (an apparatus that creates the reference image 130) via the network 91.

  The imaging unit 19 is a general digital camera, and images the printed matter 90 with a photoelectric conversion element (not shown) to acquire image data (FIG. 3: target image 131). That is, the imaging unit 19 has a function corresponding to the acquisition unit of the present invention. In addition, as a structure which acquires the target image 131 from printed matter, a scanner etc. may be sufficient, for example.

  FIG. 3 is a diagram illustrating functional blocks of the defect detection apparatus 1 together with a data flow. The first generation unit 100, the coarse image generation unit 101, the second generation unit 102, the enlargement unit 103, and the defect detection unit 104 illustrated in FIG. 3 are functional blocks that are realized by the CPU 10 operating according to the program 120.

  The first generation unit 100 generates a difference image 132 that is a difference between them based on the reference image 130 and the target image 131.

  The coarse image generation unit 101 generates a reference coarse image 133 obtained by reducing the reference image 130 based on the reference image 130. The coarse image generating unit 101 generates a target coarse image 134 obtained by reducing the target image 131 based on the target image 131. Note that the method of reducing the reference image 130 and the target image 131 by the coarse image generation unit 101 uses a so-called bilinear method, not a reduction method by simple thinning.

  Based on the reference rough image 133 and the target rough image 134, the second generation unit 102 generates a difference rough image 135 that is a difference between them.

  The enlargement unit 103 enlarges the difference coarse image 135 to the size of the difference image 132 to generate a mask image 136. Note that the enlargement unit 103 enlarges the differential coarse image 135 using a bilinear method instead of an enlargement method based on simple enlargement.

  The defect detection unit 104 detects a defect based on the masked difference image 132 while masking the difference image 132 with the mask image 136. Further, the defect detection unit 104 causes the display unit 16 to display the detection result.

  The above is the description of the configuration and function of the defect detection apparatus 1. Next, the operation of the defect detection apparatus 1 will be described.

  FIG. 4 is a flowchart showing the operation of the defect detection apparatus 1. When the power is turned on, the defect detection device 1 performs a predetermined initial setting and then enters a standby state until an operator instruction is input (step S1).

  When the operator operates the operation unit 15 to input an instruction to the defect detection apparatus 1 so as to start detecting defects in the printed matter 90, the CPU 10 determines Yes in step S1 and executes an acquisition process (step S2). ).

  In the acquisition process, first, the reference image 130 is read from the medium 92 inserted in the drive device 18, transferred to the hard disk 14 of the storage unit 11, and stored. Further, the reference image 130 is acquired by reading the reference image 130 from the hard disk 14 into the RAM 13.

  Further, in the acquisition process, the printed material 90 is imaged by the imaging unit 19, and the target image 131 with the printed material 90 as a subject is transferred to the RAM 13. Thereby, the target image 131 is acquired.

  FIG. 5 is a diagram illustrating the reference image 130. FIG. 6 is a diagram illustrating the target image 131. In the example shown here, the roof color in the target image 131 is printed in a darker color than the roof color in the reference image 130. That is, there is a difference in shading between the reference image 130 and the target image 131 in the color of the roof. In addition, the target image 131 has a contour 131 a that is not in the reference image 130.

  Once the reference image 130 and the target image 131 are acquired, the first generation process is then executed (step S3).

  In the first generation step, the first generation unit 100 extracts corresponding pixels (pixels having the same position in each image) in the reference image 130 and the target image 131, and the pixel values of the two extracted pixels are obtained. Calculate the difference. Furthermore, a difference image 132 is generated in which the difference between the obtained pixel values is a pixel value.

  FIG. 7 is a diagram illustrating the difference image 132. Note that points A, B, C, and D in FIG. 7 are given for the following explanation. In the difference image 132, the pixel value of the pixel corresponding to the region where the reference image 130 and the target image 131 match is “0”.

  Therefore, as shown in FIG. 7, in the difference image 132 in the example shown here, a mismatch area 132a caused by the presence / absence of the contour 131a and a mismatch area 132b caused by the difference in shade of the roof color are expressed.

  FIG. 8 is a diagram illustrating pixel values of pixels from the point A to the point B of the difference image 132 illustrated in FIG. FIG. 9 is a diagram illustrating pixel values of pixels from point C to point D of the difference image 132 illustrated in FIG. 7. Here, it is assumed that the pixel value of the pixel in the mismatch area 132a is equal to the pixel value of the pixel in the mismatch area 132b.

  As is apparent from FIG. 8, in the difference image 132, only the pixel corresponding to the contour 131a in the target image 131 has a high pixel value. As is clear from FIG. 9, in the difference image 132, only the pixel corresponding to the “roof” in the reference image 130 (and the target image 131) has a high pixel value.

  When the difference image 132 is generated, the defect detection apparatus 1 executes a coarse image generation process (step S4).

  In the coarse image generation process, first, the coarse image generation unit 101 performs a reduction process by a bilinear method on the reference image 130 to generate a reference coarse image 133. In addition, the coarse image generation unit 101 similarly performs a reduction process by the bilinear method on the target image 131 to generate a target coarse image 134. Note that the reduction ratio when generating the reference rough image 133 and the target rough image 134 is set and stored in advance according to how many inconsistent regions are to be detected as defects.

  Once the reference coarse image 133 and the target coarse image 134 are generated, the second generation process is then performed (step S5).

  In the second generation step, the second generation unit 102 generates a differential coarse image 135 based on the reference coarse image 133 and the target coarse image 134. At this time, the second generation unit 102 generates a differential coarse image 135 while performing a so-called “blurring process” using a blur mask of a predetermined size.

  As described above, the second generation unit 102 performs the blurring process when generating the differential coarse image 135, thereby causing a mismatch area between the reference coarse image 133 and the target coarse image 134 (that is, the reference image 130 and the target image). The pixel value can be further reduced for a pixel in a relatively small area of the non-matching area 131.

  FIG. 10 is a diagram illustrating the differential coarse image 135. Due to the reduction processing by the bilinear method and the blurring processing, in the difference rough image 135, the pixel value (difference value) of a pixel corresponding to a relatively small region of the mismatched region between the reference image 130 and the target image 131 is a low value. (Approximate to "0").

  When the difference rough image 135 is generated, the defect detection apparatus 1 executes an enlargement process (step S6).

  In the enlargement process, the enlargement unit 103 enlarges the differential coarse image 135 by the bilinear method to generate a mask image 136. By this enlargement process, the difference coarse image 135 becomes the same size as the difference image 132.

  FIG. 11 is a diagram illustrating a mask image 136. 12 is a diagram showing pixel values of pixels from point A to point B shown in FIG. 7 in the mask image 136 shown in FIG. FIG. 13 is a diagram showing pixel values of pixels from point C to point D shown in FIG. 7 in the mask image 136 shown in FIG.

  Although not recognized in FIG. 11, as shown in FIG. 12, the pixel value of the mask image 136 does not become “0” even for the mismatch region caused by the contour 131 a in the target image 131. However, as apparent from a comparison between FIG. 8 and FIG. 12, in the mask image 136, the pixel value of the mismatch region caused by the contour 131 a in the target image 131 is greatly reduced.

  On the other hand, as apparent from FIG. 11, the area around the mismatch area 136 a caused by the difference in shade of the roof color is slightly blurred (that is, the difference value is lowered). However, as is clear from a comparison between FIG. 9 and FIG. 13, the pixel value of the inner region of the mismatch region 136a is not lowered as compared to the mismatch region 132b.

  When the mask image 136 is generated, the defect detection apparatus 1 performs a defect detection process (step S7).

  In the defect detection step, first, the defect detection unit 104 masks the difference image 132 using the mask image 136. That is, the pixel value of the corresponding pixel of the mask image 136 is subtracted from the pixel value of each pixel of the difference image 132 (a negative value is “0”) to generate an inspection image.

  FIG. 14 is a diagram illustrating an image for inspection. FIG. 15 is a diagram illustrating pixel values of pixels from point A to point B shown in FIG. 7 in the inspection image shown in FIG. FIG. 16 is a diagram showing pixel values of pixels from point C to point D shown in FIG. 7 in the inspection image shown in FIG.

  When the inspection image is generated, the defect detection unit 104 compares the pixel value of the image with a predetermined threshold value, and determines that the area having a pixel value (difference value) that is greater than or equal to the allowable range is “defect” Detect as.

  As apparent from a comparison between FIG. 8 and FIG. 9, in the difference image 132, the mismatch area 132a caused by the contour 131a and the mismatch area 132b caused by the difference in shade of the roof color have the same pixel value. Therefore, no matter how the threshold value is set, only the mismatch area 132a cannot be detected as a defect.

  That is, as described above, if a threshold value is set to detect a region such as the non-matching region 132a as a defect, a region such as the non-matching region 132b (for example, a density difference consciously given by the operator is generated). Region) is also erroneously detected as a defect. If a threshold value is set so as not to erroneously detect the inconsistent area 132b, an area such as the inconsistent area 132a is overlooked, and the detection accuracy decreases.

  However, as apparent from comparison between FIG. 15 and FIG. 16, in the inspection image used for defect detection in the defect detection apparatus 1, the pixel value of the inconsistent region 137 a caused by the contour 131 a is the shade of the roof color. The pixel value is higher than the pixel value in the non-matching region due to the difference.

  Therefore, by setting an appropriate threshold value, the defect detection unit 104 detects the inconsistent area 137a as a defect, while erroneously detecting the inconsistent area due to a local ink unevenness such as a difference in shade of the roof color as a defect. Can be suppressed.

  Further, the defect detection unit 104 displays the detection result on the display unit 16. As the detection result, for example, an image obtained by binarizing the inspection image shown in FIG. 14 with a predetermined threshold value is displayed on the display unit 16.

  When the defect detection process is finished, it is further determined whether or not the detection process is finished (step S8). If the detection process is to be continued, the process returns to step S1 to continue the process. If the detection process is to be terminated, the process is terminated.

  As described above, the defect detection apparatus 1 according to the present embodiment generates the difference image 132 based on the reference image 130 and the target image 131, and generates the difference rough image 135 based on the reference rough image 133 and the target rough image 134. A mask image 136 is generated by enlarging the generated difference rough image 135 to the size of the difference image 132, and the generated mask image 136 masks the difference image 132 to detect defects. It is possible to inspect while removing the influence of uneven ink. Accordingly, it is possible to prevent erroneous detection related to relatively gentle ink unevenness that is not recognized as a defect by a viewer without reducing the detection accuracy of the defect.

<2. Second Embodiment>
In the first embodiment, in order to generate a differential coarse image, a reference coarse image obtained by reducing the reference image and a target coarse image obtained by reducing the target image are generated. However, the method for generating the differential coarse image is not limited to this.

  FIG. 17 is a diagram illustrating functional blocks of the defect detection apparatus 1 according to the second embodiment together with a data flow.

  The defect detection apparatus 1 in the second embodiment does not include a functional block corresponding to the coarse image generation unit 101 of the defect detection apparatus 1 in the first embodiment. Moreover, the 2nd production | generation part 102a is provided as a structure similar to the 2nd production | generation part 102 of the defect detection apparatus 1 in 1st Embodiment.

  Note that in the defect detection device 1 according to the second embodiment, components having substantially the same functions as those of the defect detection device 1 according to the first embodiment are denoted by the same reference numerals, and description thereof is omitted as appropriate. The same applies to the following embodiments.

  The second generation unit 102a in the second embodiment generates the difference coarse image 135 by reducing the difference image 132 generated by the first generation unit 100 by the bilinear method. It differs from the 2nd production | generation part 102 in the form. That is, in the defect detection device 1 according to the second embodiment, data corresponding to the reference coarse image 133 and the target coarse image 134 is not generated.

  The operation of the defect detection apparatus 1 in the second embodiment will be described. In the defect detection apparatus 1 in the second embodiment, a process corresponding to the coarse image generation process (step S4) in the first embodiment is as follows. Not executed.

  That is, after the first generation process (step S3) is executed, the second generation process (step S5) is executed. However, in the second generation step in the second embodiment, the second generation unit 102 a generates the difference coarse image 135 based on the difference image 132.

  The other steps are the same as the operation of the defect detection apparatus 1 in the first embodiment, and therefore the description thereof is omitted below.

  As described above, the defect detection device 1 according to the second embodiment generates the difference coarse image 135 based on the difference image 132, thereby obtaining the same effect as the defect detection device 1 according to the first embodiment. Obtainable.

<3. Third Embodiment>
In the above embodiment, for example, as shown in FIG. 16, pixels having a relatively high pixel value appear in the inspection image even if there is a general ink unevenness. Even in this case, as described above, if an appropriate threshold is determined, it is possible to determine whether or not the defect is present. However, if a pixel having a relatively high pixel value appears due to a phenomenon that is not desired to be detected as a defect, an appropriate range of a preset threshold value is narrowed. That is, there is a problem that it is difficult to set an appropriate threshold value. The present invention also provides a solution for such problems.

  FIG. 18 is a diagram illustrating functional blocks of the defect detection apparatus 1 according to the third embodiment together with a data flow.

  The defect detection device 1 according to the third embodiment is that the functional block similar to the enlargement unit 103 of the defect detection device 1 according to the second embodiment is an enlargement unit 103a having a thickening processing unit 105. This is different from the defect detection apparatus 1 in the second embodiment.

  The enlargement unit 103a generates an image (not shown) corresponding to the mask image 136 in the second embodiment, based on the difference coarse image 135, similarly to the enlargement unit 103 in the second embodiment. . Further, a thickening process is performed on the image generated in this way by the thickening processing unit 105 to generate a mask image 138.

  The fattening process is a process for obtaining the maximum pixel value of a predetermined number of adjacent pixels (for example, 3 × 3 pixels) and setting all the pixel values of the predetermined number of pixels to the maximum pixel value obtained. is there.

  FIG. 19 is a diagram illustrating a mask image 138. The mask image 138 shown in FIG. 19 is generated by the thickening processing unit 105 executing the thickening process on the same image as the mask image 136 illustrated in FIG.

  20 is a diagram showing pixel values of pixels from point A to point B shown in FIG. 7 in the mask image 138 shown in FIG. In addition, the broken line shown in FIG. 20 shows the pixel value shown in FIG. That is, the broken line shown in FIG. 20 indicates the pixel value when the fattening process is not performed.

  Further, FIG. 21 is a diagram showing pixel values of pixels from point C to point D shown in FIG. 7 in the mask image 138 shown in FIG. The broken line shown in FIG. 21 indicates the pixel value shown in FIG. That is, the broken line shown in FIG. 21 indicates the pixel value when the fattening process is not performed.

  As is apparent from FIGS. 20 and 21, when the fattening process is performed, the maximum value of the pixel value in the mismatch region does not change, but an image in which the periphery of the actual mismatch region is spread is obtained.

  In the defect detection step corresponding to step S6, the defect detection unit 104 according to the second embodiment detects defects while masking the difference image 132 with the thickened mask image 138 generated by the enlargement unit 103a. I do.

  FIG. 22 is a diagram illustrating pixel values of pixels from point A to point B illustrated in FIG. 7 in the inspection image according to the third embodiment. FIG. 23 is a diagram illustrating pixel values of pixels from point C to point D illustrated in FIG. 7 in the inspection image according to the third embodiment.

  As apparent from a comparison between FIG. 22 and FIG. 23, the pixel value of the mismatched region caused by the contour 131a is sufficiently higher than the pixel value of the mismatched region caused by the shade color difference. ing. In other words, as apparent from a comparison between FIG. 16 and FIG. 23, the pixel value of the non-matching region due to the difference in shade of the roof color is almost “0”. That is, since the mismatch due to the global ink unevenness can be effectively eliminated, the threshold value can be easily set in the defect detection unit 104.

  In the present embodiment, the example in which the fat processing unit 105 is provided in the defect detection apparatus 1 in the second embodiment has been described. However, the fat processing unit 105 in the defect detection apparatus 1 in the first embodiment is described. May be provided.

  Further, in the present embodiment, it has been described that the thickening process is performed on the enlarged differential coarse image. However, the thickening process may be performed on the differential image 132 or the differential coarse image 135, for example.

<4. Modification>
Although the embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and various modifications can be made.

  For example, in the above-described embodiment, the processing for enlarging / reducing an image has been described as using the bilinear method. However, the present invention is not limited to this, and a bicubic method may be used. In other words, the present invention can be applied to image enlargement / reduction processing according to the present invention unless it is reduced by simple decimation or enlargement by simple duplication.

  Moreover, each process shown to the said embodiment is an illustration, Comprising: It is not limited to this. That is, as long as the same effect can be obtained, the operation order and the operation content may be changed as appropriate.

  In the above embodiment, each functional block has been described as being realized by software by the program 120. However, some or all of these functional blocks may be realized by hardware using a dedicated circuit. Good.

FIG. 1 is a diagram showing a defect detection apparatus according to the present invention. It is a figure which shows the structure of a defect detection apparatus. It is a figure which shows the functional block of the defect detection apparatus in 1st Embodiment with the flow of data. It is a flowchart which shows operation | movement of the defect detection apparatus in 1st Embodiment. It is a figure which illustrates a reference image. It is a figure which illustrates an object picture. It is a figure which illustrates a difference image. It is a figure which shows the pixel value of the pixel from the point A to the point B of the difference image shown in FIG. It is a figure which shows the pixel value of the pixel from the point C of the difference image shown in FIG. 7 to the point D. It is a figure which illustrates a difference rough image. It is a figure which illustrates the image for masks. FIG. 12 is a diagram illustrating pixel values of pixels from point A to point B shown in FIG. 7 in the mask image shown in FIG. 11. FIG. 12 is a diagram illustrating pixel values of pixels from a point C to a point D illustrated in FIG. 7 in the mask image illustrated in FIG. 11. It is a figure which illustrates the image for inspection. It is a figure which shows the pixel value of the pixel from the point A shown in FIG. 7 to the point B in the image for an inspection shown in FIG. It is a figure which shows the pixel value of the pixel from the point C shown in FIG. 7 to the point D in the image for an inspection shown in FIG. It is a figure which shows the functional block of the defect detection apparatus in 2nd Embodiment with the flow of data. It is a figure which shows the functional block of the defect detection apparatus in 3rd Embodiment with the flow of data. It is a figure which illustrates the image for masks. FIG. 20 is a diagram showing pixel values of pixels from point A to point B shown in FIG. 7 in the mask image shown in FIG. 19. FIG. 20 is a diagram illustrating pixel values of pixels from a point C to a point D illustrated in FIG. 7 in the mask image illustrated in FIG. 19. It is a figure which shows the pixel value of the pixel from the point A shown in FIG. 7 to the point B in the image for a test | inspection in 3rd Embodiment. It is a figure which shows the pixel value of the pixel from the point C shown in FIG. 7 to the point D in the image for a test | inspection in 3rd Embodiment.

Explanation of symbols

1 Defect detection device 10 CPU
DESCRIPTION OF SYMBOLS 100 1st production | generation part 101 Coarse image generation part 102,102a 2nd production | generation part 103,103a Enlargement part 104 Defect detection part 105 Thickening process part 11 Storage part 120 Program 130 Reference image 131 Target image 132 Difference image 133 Reference rough image 134 Target coarse image 135 Differential coarse image 136, 138 Mask image 18 Drive device (acquisition means)
19 Imaging unit (acquisition means)
90 printed matter

Claims (6)

A defect detection device for detecting defects in a target image,
An acquisition means for acquiring a reference image and a target image;
First generation means for generating a difference image based on the reference image and the target image acquired by the acquisition means;
A second generating means for generating a differential coarse image;
Enlargement means for generating a mask image by enlarging the difference coarse image generated by the second generation means to the size of the difference image generated by the first generation means;
A defect detection means for detecting a defect by masking the difference image generated by the second generation means with the mask image generated by the enlargement means;
A defect detection apparatus comprising: The defect detection apparatus according to claim 1,
A rough image generation unit that generates a reference coarse image based on the reference image acquired by the acquisition unit and that generates a target coarse image based on the target image acquired by the acquisition unit,
The second generation means includes
A defect detection apparatus, characterized in that a differential coarse image is generated based on a reference coarse image and a target coarse image generated by the coarse image generation means. The defect detection apparatus according to claim 1,
The defect detection apparatus, wherein the second generation unit generates a differential coarse image based on the difference image generated by the first generation unit. The defect detection apparatus according to any one of claims 1 to 3,
A defect detection apparatus, wherein the mask image is subjected to a fattening process. A defect detection method for detecting defects in a target image,
An acquisition step of acquiring a reference image and a target image;
A first generation step of generating a difference image based on the reference image and the target image acquired in the acquisition step;
A second generation step of generating a differential coarse image;
An enlargement step of generating a mask image by enlarging the difference coarse image generated in the second generation step to the size of the difference image generated in the first generation step;
A defect detection step of detecting a defect by masking the difference image generated in the second generation step by the mask image generated in the enlargement step;
A defect detection method comprising: A computer-readable program, wherein execution of the program by the computer causes the computer to
An acquisition means for acquiring a reference image and a target image;
First generation means for generating a difference image based on the reference image and the target image acquired by the acquisition means;
A second generating means for generating a differential coarse image;
Enlargement means for generating a mask image by enlarging the difference coarse image generated by the second generation means to the size of the difference image generated by the first generation means;
A defect detection means for detecting a defect by masking the difference image generated by the second generation means with the mask image generated by the enlargement means;
With
A program that functions as a defect detection device that detects a defect in a target image.

Images