JP2008014703A - Image flaw inspection system, image flaw inspection method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for inspecting another part with high flaw detection sensitivity, while preventing the edge part of an image from being detected as flaw, in an image flaw inspection system. <P>SOLUTION: The image flaw inspection system 40 is constituted so as to set a margin value at each pixel on the basis of the edge image data D5 formed from reference image data D1, and the flaws of the image are inspected by using the margin value different at each pixel. Accordingly, for determining the presence of the flaw, margin values different in the edge part A3 and other regions A1 and A2 are used. As a result, the other parts A1 and A2 can be inspected with high flaw detection sensitivity, while preventing the edge part A3 from being detected as a flaw. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image defect inspection technique for performing a defect inspection on a target image by comparing a reference image and the target image.

  Conventionally, in a printing process, it is inspected whether an image printed on a printed material is free from defects. When performing such an image defect inspection, first, an image on a printed matter is read by an imaging device such as a CCD scanner to obtain an inspection target image. Then, a difference image between the inspection target image and the reference image is generated, and the density value of each pixel of the difference image is compared with a specified margin value to determine the presence / absence of a defect in each pixel.

  For example, Patent Literature 1 discloses a conventional technique related to image defect inspection of printed matter.

JP-A-10-19796

  By the way, when an edge portion (a portion where the density value changes greatly) exists in the image on the printed matter, when the image is read by a CCD scanner, an intermediate density band (aliasing) occurs in the edge portion, The position may be slightly shifted. In most cases, such an edge error has no problem on the printed matter, and it is desirable not to detect it as a defect in the image defect inspection.

  However, in the conventional image defect inspection, if the margin value is set to be small, the error of the edge portion in the inspection target image may be detected as a large number of defects. If the margin value is set to be large, the number of detected defects in such an edge portion is reduced. However, if this is done, the defect detecting power for the region other than the edge portion is also lowered, which is not desirable.

  The present invention has been made in view of such circumstances, and inspecting other parts with high defect detection sensitivity while preventing the edge part of the image from being detected as a defect in the image defect inspection. It aims at providing the technology that can do.

  In order to solve the above-described problem, the invention according to claim 1 is an image defect inspection apparatus that performs a defect inspection on the target image by comparing the reference image and the target image, and an edge whose density is switched in the reference image. Edge image generating means for generating an edge image by extracting a part; margin value setting means for setting a margin value for each pixel based on the edge image; and generating a difference image between the reference image and the target image Differential image generation means for performing the determination, and defect determination means for determining the presence or absence of a defect for each pixel by comparing the value of each pixel of the difference image with the margin value of the pixel. .

  The invention according to claim 2 is the image defect inspection apparatus according to claim 1, wherein the edge image generation unit is configured to change the edge image in the edge image according to a degree of density change of the edge portion in the reference image. The density value of the edge portion is set in multiple stages.

  The invention according to claim 3 is the image defect inspection apparatus according to claim 1 or 2, wherein the edge image generation means performs a fattening process on the edge portion of the edge image. To do.

  The invention according to claim 4 is the image defect inspection apparatus according to any one of claims 1 to 3, wherein the margin value setting means calculates a correction coefficient for each pixel based on the edge image. The margin value of each pixel is set based on a standard margin value and the correction coefficient.

  The invention according to claim 5 is the image defect inspection apparatus according to claim 4, wherein the margin value setting unit is configured to determine each pixel based on a value of each pixel of the edge image and a predetermined lookup table. The correction coefficient is calculated.

  The invention according to claim 6 is an image defect inspection method for performing a defect inspection on the target image by comparing a reference image with a target image, and extracting an edge portion where the density is switched in the reference image. An edge image generation step for generating an image; a margin value setting step for setting a margin value for each pixel based on the edge image; a difference image generation step for generating a difference image between the reference image and the target image; A defect determination step of comparing each pixel value of the difference image with the margin value of the pixel to determine the presence / absence of a defect for each pixel.

  The invention according to claim 7 is a program for performing a defect inspection on the target image by comparing the reference image and the target image, and extracting, to a computer, an edge portion whose density is switched in the reference image. An edge image generation procedure for generating an edge image by using a margin value setting procedure for setting a margin value for each pixel based on the edge image, and a difference image generation procedure for generating a difference image between the reference image and the target image And a defect determination procedure for determining the presence / absence of a defect for each pixel by comparing the value of each pixel of the difference image with the margin value of the pixel.

  The invention according to claim 8 is the program according to claim 7, wherein the edge image generation procedure is performed according to a degree of density change of the edge portion in the reference image. The density value is set in multiple stages.

  The invention according to claim 9 is the program according to claim 7 or 8, wherein the edge image generation procedure performs a fattening process on the edge portion of the edge image.

  The invention according to claim 10 is the program according to any one of claims 7 to 9, wherein the margin value setting procedure calculates a correction coefficient of each pixel based on the edge image, and The margin value of each pixel is set based on a typical margin value and the correction coefficient.

  The invention according to claim 11 is the program according to claim 10, wherein the margin value setting procedure is based on a value of each pixel of the edge image and a predetermined lookup table, and a correction coefficient for each pixel. Is calculated.

  According to the first to fifth aspects of the present invention, the image defect inspection apparatus includes an edge image generation unit that generates an edge image by extracting an edge portion whose density is switched in the reference image, and each pixel based on the edge image. By comparing the margin value setting means for setting the margin value, the difference image generation means for generating a difference image between the reference image and the target image, the value of each pixel of the difference image and the margin value of the pixel, Defect determination means for determining the presence or absence of a defect for each pixel. For this reason, different margin values can be set for the edge portion and other regions, and the presence or absence of defects can be determined based on the different margin values for each region. Therefore, it is possible to inspect other portions with high defect detection sensitivity while preventing the edge portion from being detected as a defect.

  In particular, according to the second aspect of the invention, the edge image generation means sets the density value of the edge portion in the edge image in multiple stages according to the degree of density change of the edge portion in the reference image. Therefore, an optimum margin value can be set according to the degree of density change of the edge portion in the reference image.

  In particular, according to the invention described in claim 3, the edge image generation means performs the fattening process on the edge portion in the edge image. For this reason, the area to which the margin value corresponding to the edge portion is applied is enlarged. Therefore, the margin value corresponding to the edge portion can be reliably applied to the edge portion.

  In particular, according to the fourth aspect of the present invention, the margin value setting means calculates a correction coefficient for each pixel based on the edge image, and the margin value for each pixel based on the standard margin value and the correction coefficient. Set. For this reason, the margin value of each pixel can be increased or decreased based on the standard margin value.

  In particular, according to the invention described in claim 5, the margin value setting means calculates the correction coefficient of each pixel based on the value of each pixel of the edge image and a predetermined lookup table. For this reason, the relationship between the value of each pixel of the edge image and the correction coefficient can be set freely and finely. Therefore, an optimal correction coefficient can be calculated according to various edges.

  According to the invention described in claim 6, the image defect inspection method includes an edge image generation step of generating an edge image by extracting an edge portion whose density is switched in the reference image, and a pixel-by-pixel basis based on the edge image. By comparing the margin value setting step for setting a margin value to the difference image generation step for generating a difference image between the reference image and the target image, and the value of each pixel of the difference image and the margin value of the pixel, A defect determination step for determining the presence or absence of a defect for each pixel. For this reason, different margin values can be set for the edge portion and other regions, and the presence or absence of defects can be determined based on the different margin values for each region. Therefore, it is possible to inspect other portions with high defect detection sensitivity while preventing the edge portion from being detected as a defect.

  According to the invention described in claims 7 to 11, the program is based on an edge image generation procedure for generating an edge image by extracting an edge portion whose density is switched in the reference image, and the edge image. The margin value setting procedure for setting a margin value for each pixel, the difference image generation procedure for generating a difference image between the reference image and the target image, and the value of each pixel of the difference image and the margin value of the pixel are compared. Thus, a defect determination procedure for determining the presence or absence of a defect for each pixel is executed. For this reason, different margin values can be set for the edge portion and other regions, and the presence or absence of defects can be determined based on the different margin values for each region. Therefore, it is possible to inspect other portions with high defect detection sensitivity while preventing the edge portion from being detected as a defect.

  In particular, according to the eighth aspect of the invention, the edge image generation procedure sets the density value of the edge part in the edge image in multiple stages according to the degree of density change of the edge part in the reference image. Therefore, an optimum margin value can be set according to the degree of density change of the edge portion in the reference image.

  In particular, according to the invention described in claim 9, the edge image generation procedure performs the fattening process on the edge portion in the edge image. For this reason, the area to which the margin value corresponding to the edge portion is applied is enlarged. Therefore, the margin value corresponding to the edge portion can be reliably applied to the edge portion.

  In particular, according to the invention described in claim 10, the margin value setting procedure calculates a correction coefficient for each pixel based on the edge image, and the margin value for each pixel based on the standard margin value and the correction coefficient. Set. For this reason, the margin value of each pixel can be increased or decreased based on the standard margin value.

  In particular, according to the invention described in claim 11, the margin value setting procedure calculates the correction coefficient of each pixel based on the value of each pixel of the edge image and a predetermined look-up table. For this reason, the relationship between the value of each pixel of the edge image and the correction coefficient can be set freely and finely. Therefore, an optimal correction coefficient can be calculated according to various edges.

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

<1. Configuration of printing system>
FIG. 1 is a block diagram illustrating a configuration of a printing system 1 including an image defect inspection apparatus 40 according to an embodiment of the present invention. The printing system 1 performs image processing on the input image, creates and prints a printing plate for the image after image processing, and further performs defect inspection on the image on the created printed matter. System. As shown in FIG. 1, the printing system 1 includes an image processing apparatus 10, a plate making apparatus 20, a printing apparatus 30, and an image defect inspection apparatus 40.

  The image processing apparatus 10 is an apparatus for converting input image data into image data for printing. The image processing apparatus 10 receives image data described in the format of Portable Document Format or PostScript (registered trademark of Adobe Systems), and performs image processing such as RIP processing on the image data. The image data is divided into pixels according to the resolution at the time of output, and becomes image data (bitmap data) in which the density is expressed by the density value of each pixel. Since the image data generated by this image processing is image data that becomes a reference for output in the subsequent printing processing, it is hereinafter referred to as reference image data D1. The reference image data D1 generated in the image processing apparatus 10 is transmitted to the plate making apparatus 20 through the communication line 51 and is transmitted to the image defect inspection apparatus 40 through the communication line 52.

  The plate making apparatus 20 is an apparatus for creating the printing plate P1 based on the reference image data D1 received from the image processing apparatus 10. The plate making apparatus 20 creates the printing plate P1 by recording the reference image data D1 on the metal plate for each color element. For example, the plate making apparatus 20 is configured to include a drum and a recording head, and irradiates the metal plate with a laser beam from the recording head while rotating the metal plate coated with a photosensitive material on the surface of the drum. By doing so, a reference image is recorded on the metal plate.

  The printing apparatus 30 is an apparatus for printing a reference image on printing paper using the printing plate P1 created in the plate making apparatus 20. For example, the printing apparatus 30 performs printing by mounting the printing plate P1 on a cylindrical plate cylinder and transferring the image from the printing plate P1 to the printing paper while conveying the printing paper. The printed matter P2 created as a result of printing is delivered to the image defect inspection apparatus 40. Since the image printed on the printed material P2 is an image to be inspected in the image defect inspection apparatus 40, it is hereinafter referred to as an inspection object image I.

  The image defect inspection apparatus 40 is an apparatus for reading the inspection target image I on the printed material P2 and performing the defect inspection of the inspection target image I. As shown in FIG. 1, the image defect inspection apparatus 40 includes an image data acquisition unit 41 and a defect inspection processing unit 42. The image data acquisition unit 41 images the inspection target image I on the printed material P2 by the imaging unit 41b while scanning the printed material P2 vertically and horizontally by the scanning drive unit 41a. The imaging unit 41b reads the inspection target image I with an imaging element such as a CCD element, and acquires inspection target image data D2 including a plurality of pixels. The acquired inspection target image data D2 is transmitted from the imaging unit 41b to the defect inspection processing unit 42.

  The defect inspection processing unit 42 performs defect inspection of the inspection target image data D2 by comparing the reference image data D1 received from the image processing apparatus 10 with the inspection target image data D2 received from the imaging unit 41b. FIG. 2 is a block diagram showing a detailed configuration of the defect inspection processing unit 42. As shown in FIG. 2, the defect inspection processing unit 42 includes an edge image generation unit 42a, a correction coefficient calculation unit 42b, a margin value setting unit 42c, a difference image generation unit 42d, and a defect determination unit 42e. Have. The defect inspection processing unit 42 is specifically configured by a computer including a CPU, a memory, a hard disk, and the like. The operation of each unit described above is performed based on a program 42g installed in the computer. It is realized by doing.

  The edge image generation unit 42a is a processing unit for extracting a portion (edge portion) where the density value is largely switched in the reference image data D1 and generating the edge image data D5. The processing of the edge image generation unit 42a will be described with reference to the example of FIG. The reference image data D1 of FIG. 3 has a low density value area A1 and a high density value area A2, and the boundary between the areas A1 and A2 is an edge portion A3 where the density value is largely switched. When such reference image data D1 is input to the edge image generation unit 42a, the edge image generation unit 42a first performs a blurring process on the reference image data D1. That is, the edge image generation unit 42a generates the intermediate density band B1 of the areas A1 and A2 at the edge portion A3 of the reference image data D1, and generates blurred image data D3.

  Next, the edge image generation unit 42a generates difference image data D4 by taking the difference between the reference image data D1 and the blurred image data D3. Then, the edge image generation unit 42a generates edge image data D5 by performing a fattening process on the difference image data D4. The edge image data D5 is image data composed of a plurality of multi-value pixels, and the density value of the edge portion in the edge image data D5 reflects the degree of density change of the edge portion A3 in the reference image data D1. It has become.

  The correction coefficient calculation unit 42b is a processing unit that calculates a correction coefficient for correcting a standard margin value according to an image. The processing of the correction coefficient calculation unit 42b will be described with reference to the example of FIG. As shown in FIG. 4, the correction coefficient calculation unit 42b performs pixel detection based on the density value of each pixel of the edge image data D5 generated by the edge image generation unit 42a and a preset lookup table LUT. A correction coefficient is calculated for each. As a result, the correction coefficient c1 is set for the edge portion A3, the correction coefficient c2 is set for the other areas, and different correction coefficients are set for each area. The look-up table LUT is stored in the defect inspection processing unit 42 as a set of conversion coefficients for converting the density value of each pixel in the edge image data D5 into a correction coefficient. The conversion coefficient in the lookup table LUT can be arbitrarily changed by operating a predetermined input unit (not shown).

  The margin value setting unit 42c is a processing unit for correcting a standard margin value using the correction coefficient calculated by the correction coefficient calculation unit 42b and setting an image-specific margin value. The processing of the margin value setting unit 42c will be described with reference to the example of FIG. As shown in FIG. 5, the margin value setting unit 42c multiplies the standard margin value and the correction coefficient of each pixel to set a unique margin value for each pixel. In the example of FIG. 5, a specific margin value m2 is set in the edge portion A3 by multiplying the standard margin value m1 and the correction coefficient c1. In other regions, a unique margin value m3 is set by multiplying the standard margin value m1 and the correction coefficient c2. The margin value m2 set in the edge portion A3 reflects the degree of density change of the edge portion A3 in the reference image data D1.

  The difference image generation unit 42d is a processing unit for generating difference image data D6 by taking the difference between the reference image data D1 and the inspection target image data D2. The process of the difference image generation unit 42d will be described with reference to the example of FIG. The difference image generation unit 42d compares the reference image data D1 and the inspection target image data D2 for each pixel, and calculates the absolute value of the difference in density value for each pixel. Then, the difference image data D6 is configured with the calculated value as the density value of each pixel. As shown in FIG. 5, in the difference image data D6, relatively high density values remain in the defective portions A4 and A5 and the edge portion A3.

  The defect determination unit 42e uses the margin value set in the margin value setting unit 42c and the difference image data D6 generated in the difference image generation unit 42d to determine the presence / absence of a defect for each pixel. It is. Processing of the defect determination unit 42e will be described with reference to the example of FIG. The defect determination unit 42e compares the density value of each pixel of the difference image data D6 with the margin value corresponding to the pixel. If the density value of the pixel is equal to or less than the margin value, the defect determination unit 42e sets the pixel as a normal pixel. judge. Further, when the density value of the pixel exceeds the margin value, the defect determination unit 42e determines the pixel as a defective pixel.

  Here, the margin value setting unit 42c sets a particularly large margin value in the edge portion A3. That is, the inspection sensitivity of the edge portion A3 is partially reduced. For this reason, the defect determination unit 42e does not determine the density value remaining in the edge portion A3 as a large number of defects, and determines only the defect portions A4 and A5 as defects. The edge image generation unit 42a thickens the edge portion A3 of the edge image data D5 by the thickening process. For this reason, the region where the margin value is set to m2 is slightly wider than the edge portion A3. Therefore, the defect determination unit 42e can reliably apply the margin value m2 to the edge portion A3. For example, the difference image generation unit 42d performs a “swing process” in which the reference image data D1 and the inspection target image data D2 are shifted by several pixels to obtain a difference in order to allow a slight positional shift of the inspection target image D2. Even in such a case, the defect determination unit 42e can reliably apply the margin value m2 to the edge portion A3.

<2. Processing of printing system>
Next, a processing flow when the printed system P1 creates the printed matter P2 and performs the image defect inspection will be described with reference to the flowchart of FIG. First, when image data is input to the image processing apparatus 10, the image processing apparatus 10 performs image processing on the input image and generates reference image data D1 (step S1). The image processing apparatus 10 transmits the generated reference image data D1 to the plate making apparatus 20 and the defect inspection processing unit 42 of the image defect inspection apparatus 40.

  When receiving the reference image data D1 from the image processing apparatus 10, the plate making apparatus 20 creates the printing plate P1 by recording the reference image data D1 on the metal plate (step S2). The created printing plate P1 is conveyed to the printing apparatus 30. Then, the printing apparatus 30 performs printing on the printing paper using the printing plate P1, and creates a printed material P2 (step S3).

  The printed matter P2 created in the printing apparatus 30 is conveyed to the image defect inspection apparatus 40 and set in the image data acquisition unit 41 of the image defect inspection apparatus 40. The image data acquisition unit 41 acquires the inspection target image data D2 by capturing the inspection target image I on the printed matter P2 while scanning the printed matter P2 by the scan driving unit 41a (step S4). The inspection target image data D2 is transmitted from the image data acquisition unit 41 to the defect inspection processing unit 42.

  If the color space and size of the reference image data D1 and the inspection target image data D2 are different, the defect inspection processing unit 42 unifies them. Then, defect inspection of the inspection target image data D2 is performed based on the reference image data D1 and the inspection target image data D2. First, the edge image generation unit 42a generates edge image data D5 by extracting an edge portion in the reference image data D1 (step S5). The edge image data D5 is generated by taking a difference between the blurred image data D3 generated by the blurring process and the reference image data D1, and performing a fattening process on the obtained difference image data D4.

  Next, the correction coefficient calculation unit 42b calculates a correction coefficient for each pixel based on the edge image data D5 generated in step S5 (step S6). Specifically, the correction coefficient of each pixel is set based on the value of each pixel of the edge image data D5 and a predetermined lookup table LUT. Then, the margin value setting unit 42c sets a unique margin value for each pixel by multiplying the correction coefficient calculated in step S6 by the standard margin value m1 (step S7).

  On the other hand, the difference image generation unit 42d generates difference image data D6 by taking the difference between the reference image data D1 and the inspection target image data D2 (step S8). Then, the defect determination unit 42e determines the presence / absence of a defect for each pixel using the margin value set in step S7 and the difference image data D6 generated in step S8 (step S9). That is, the defect determination unit 42e compares the density value of each pixel of the difference image data D6 with the margin value corresponding to the pixel. If the density value of the pixel is equal to or less than the margin value, the defect determination unit 42e Judge as pixel. If the density value of the pixel exceeds the margin value, the pixel is determined as a defective pixel. When the defect determination process in step S9 is completed, the defect inspection processing unit 42 displays information on the detected defective pixels on the display unit 42f. Thus, a series of processes in the printing system 1 is completed.

  As described above, the image defect inspection apparatus 40 according to the present embodiment sets a margin value for each pixel based on the edge image data D5 generated from the reference image data D1. Then, image defect inspection is performed using a different margin value for each pixel. For this reason, when determining the presence or absence of a defect, different margin values are used for the edge portion A3 and the other regions A1 and A2. Therefore, it is possible to inspect the other portions A1 and A2 with high defect detection sensitivity while preventing the edge portion A3 from being detected as a defect.

<3. Modification>
As mentioned above, although one Embodiment of this invention was described, this invention is not limited to said example. For example, in the above example, the image data acquired by capturing an image on the printed matter P2 is used as the inspection target image data D2. However, the inspection target image data D2 may be digital data before printing. In particular, when an image is printed directly on printing paper by using an ink jet or the like without using a printing plate, the image data before printing is set as inspection object image data D2, and the inspection is performed by comparing with the reference image data D1. A defect on the target image data D2 may be detected.

  In the edge image generation unit 42a, the edge image data D5 is generated by performing blurring processing and fattening processing on the reference image data D1, as shown in FIG. Data D5 may be generated. FIG. 9 is a diagram showing an example of edge image data generation processing by another method. In the example of FIG. 9, the edge image generation unit 42a generates Max image data D7 by performing Max processing (processing for replacing the density value of each pixel with the maximum density value of surrounding N × N pixels) on the reference image data D1. Further, Min processing (processing for replacing the density value of each pixel with the minimum density value of surrounding N × N pixels) is performed on the reference image data D1, and Min image data D8 is generated. Then, edge image data D5 is generated by taking a difference between the Max image data D7 and the Min image data D8. As still another method, the edge image generation unit 42a may perform filter processing using a Laplacian filter or the like directly on the reference image data D1 to generate edge image data D5.

  The correction coefficient calculation unit 42b calculates the correction coefficient for each pixel using the lookup table LUT, but converts the density value of each pixel of the edge image data D5 into a correction coefficient using a predetermined function. May be. However, if the lookup table is used, the relationship between the density value and the correction coefficient can be freely defined in detail. Therefore, an optimal correction coefficient can be calculated according to various edges.

  In the above example, the correction coefficient is calculated once for each pixel, and the margin value of each pixel is calculated by multiplying the correction coefficient by a standard margin value. However, the edge image data D5 The margin value of each pixel may be calculated directly from the density value of each pixel. In this case, a conversion coefficient between the density value and the margin value of each pixel of the edge image data D5 may be set in the lookup table LUT.

It is the block diagram which showed the structure of the printing system provided with the image defect inspection apparatus. It is the block diagram which showed the detailed structure of the defect inspection process part. It is the figure which showed the example of the process by an edge image generation part. It is the figure which showed the example of the process by the correction coefficient calculation part. It is the figure which showed the example of the process by a margin value setting part. It is the figure which showed the example of the process by a difference image generation part. It is the figure which showed the example of the process by a defect determination part. 6 is a flowchart showing a flow of processing in the printing system. It is the figure which showed the example of the process of the edge image generation part which concerns on a modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Printing system 10 Image processing apparatus 20 Plate making apparatus 30 Printing apparatus 40 Defect inspection apparatus 41 Image data acquisition part 42 Defect inspection process part 42a Edge image generation part 42b Correction coefficient calculation part 42c Margin value setting part 42d Difference image generation part 42e Defect determination Part 42f display part 42g program A3 edge part A4, A5 defective part D1 reference image data D2 inspection object image data D3 blurred image data D4 difference image data D5 edge image data D6 difference image data I inspection object image LUT lookup table P1 printing plate P2 printed material c1, c2 Correction coefficient m1, m2, m3 Margin value

Claims (11)

An image defect inspection apparatus that performs a defect inspection on the target image by comparing a reference image and the target image,
Edge image generation means for generating an edge image by extracting an edge portion whose density is switched in the reference image;
Margin value setting means for setting a margin value for each pixel based on the edge image;
Difference image generation means for generating a difference image between the reference image and the target image;
Defect determination means for determining the presence or absence of a defect for each pixel by comparing the value of each pixel of the difference image with the margin value of the pixel;
An image defect inspection apparatus comprising: The image defect inspection apparatus according to claim 1,
The image defect inspection apparatus, wherein the edge image generation means sets the density value of the edge portion in the edge image in multiple stages according to the degree of density change of the edge portion in the reference image. The image defect inspection apparatus according to claim 1 or 2,
The image defect inspection apparatus, wherein the edge image generation means performs a fattening process on the edge portion of the edge image. An image defect inspection apparatus according to any one of claims 1 to 3,
The margin value setting means calculates a correction coefficient for each pixel based on the edge image, and sets the margin value for each pixel based on a standard margin value and the correction coefficient. Defect inspection equipment. The image defect inspection apparatus according to claim 4,
The margin value setting means calculates the correction coefficient of each pixel based on the value of each pixel of the edge image and a predetermined look-up table. An image defect inspection method for performing a defect inspection on the target image by comparing a reference image with the target image,
An edge image generation step of generating an edge image by extracting an edge portion whose density is switched in the reference image;
A margin value setting step for setting a margin value for each pixel based on the edge image;
A difference image generation step for generating a difference image between the reference image and the target image;
A defect determination step of determining the presence or absence of a defect for each pixel by comparing the value of each pixel of the difference image with the margin value of the pixel;
An image defect inspection method comprising: A program for performing a defect inspection on the target image by comparing a reference image and the target image,
On the computer,
An edge image generation procedure for generating an edge image by extracting an edge portion whose density is switched in the reference image;
A margin value setting procedure for setting a margin value for each pixel based on the edge image;
A difference image generation procedure for generating a difference image between the reference image and the target image;
A defect determination procedure for determining the presence or absence of a defect for each pixel by comparing the value of each pixel of the difference image with the margin value of the pixel;
A program characterized in that is executed. The program according to claim 7,
The edge image generation procedure sets the density value of the edge part in the edge image in multiple stages according to the degree of density change of the edge part in the reference image. A program according to claim 7 or claim 8, wherein
The edge image generation procedure performs a fattening process on the edge portion of the edge image. A program according to any one of claims 7 to 9,
The margin value setting procedure calculates a correction coefficient for each pixel based on the edge image, and sets the margin value for each pixel based on a standard margin value and the correction coefficient. . The program according to claim 10,
The margin value setting procedure calculates a correction coefficient of each pixel based on a value of each pixel of the edge image and a predetermined lookup table.

Images