JP2019023589A - Defect inspection system and defect inspection method - Google Patents

Abstract

A defect inspection system and a defect inspection method that make it easy to recognize the type of a defect. In a defect inspection system, a two-dimensional image F (t1) or the like whose luminance changes in a transport direction is captured by an imaging unit, and a line L1 at the same position such as the two-dimensional image F (t1) is captured by a line division processing unit. (T1) and the like are processed into image data such as a line divided image DL1 (t1) in parallel in chronological order, and the classification unit classifies the line divided images DL1 (t1) and the like having different luminances into a plurality of line divided image groups and the like. Then, the integration unit integrates the pixel values of the image data obtained by capturing the same position of the inspection target in the line division image DL1 (t1) and the like, generates image integration data and the like for each line division image group and the like, and outputs the divided output unit. , The image integrated data is divided and output according to a predetermined rule, and the appearance of the image such as the image integrated data differs for each division according to the predetermined rule, so that the type of the defect D can be easily recognized. It made. [Selection diagram] FIG.

Description

  The present invention relates to a defect inspection system and a defect inspection method.

  As a defect inspection system for inspecting a defect to be inspected based on a captured image to be inspected, for example, a defect inspection system for detecting defects such as an optical film such as a polarizing film and a retardation film, a laminated film used for a battery separator, etc. It has been known. This type of defect inspection system transports a film in the transport direction, captures a two-dimensional image of the film every discrete time, and performs defect inspection based on the captured two-dimensional image. For example, the system of Patent Document 1 divides a two-dimensional image into a plurality of lines arranged in parallel in the conveyance direction, and lines in which the same position lines in each of the two-dimensional images captured at discrete times are arranged in chronological order. Generate split images. The line division image is processed into a defect enhancement processed image in which the luminance change is enhanced. The defect enhancement processed image is integrated by integrating the pixel values of the image data of the defect enhancement processed image obtained by imaging the same position in the inspection target. Thereby, the presence or absence and position of the defect of a film are specified easily.

Japanese Patent No. 4726983

  By the way, the defects of the film include types such as protruding convex defects and recessed concave defects. In a defect inspection system, it is desirable to identify not only the presence / absence and position of a defect but also the type of defect. However, in the above technique, since the pixel values of the image data of the defect enhancement processed image are integrated with each other, the information on the protrusion and the depression of the defect disappears or cancels each other, and the defect type It is difficult to identify and improvements are desired.

  Accordingly, an object of the present invention is to provide a defect inspection system and a defect inspection method that make it easy to recognize the type of defect.

  The present invention relates to a light source that irradiates light to an inspection target, an imaging unit that captures a two-dimensional image of light that is irradiated from the light source and transmitted or reflected through the inspection target, and a light source and an imaging unit. On the other hand, a conveyance unit that conveys the inspection target relatively in the conveyance direction and an image processing unit that processes image data of the two-dimensional image captured by the imaging unit, the imaging unit includes a conveyance direction in the two-dimensional image The image processing unit captures a two-dimensional image whose luminance changes in the matching direction, the image processing unit divides the two-dimensional image into a plurality of lines arranged in parallel in the transport direction, and the two-dimensional image captured at discrete time by the imaging unit. Each of the line division processing unit that processes the image data of the line division image in which the lines at the same position in each of the lines are arranged in chronological order and the line division image processed by the line division processing unit are predetermined. For each of the classification unit that classifies into two or more line-divided image groups according to the rules and the line-divided images classified into the same line-divided image group by the classification unit, the pixel values of the image data obtained by imaging the same position in the inspection object An integration unit that generates image integration data for each line-divided image group, and a division output unit that divides and outputs the image integration data for each line-divided image group generated by the integration unit according to a predetermined rule; A defect inspection system having

  According to this configuration, the light source that irradiates the inspection target with light, the imaging unit that captures, every discrete time, a two-dimensional image of light that is irradiated from the light source and transmitted through or reflected on the inspection target, the light source and the imaging In a defect inspection system including a transport unit that transports an inspection target relative to the transport unit relative to the transport direction and an image processing unit that processes image data of a two-dimensional image captured by the imaging unit, A two-dimensional image whose luminance changes in a direction matching the conveyance direction in the two-dimensional image is captured, and the two-dimensional image is divided into a plurality of lines arranged in parallel in the conveyance direction by the line division processing unit of the image processing unit. Is processed into image data of line-divided images in which lines at the same position in each of the two-dimensional images captured at discrete times are arranged in chronological order. It becomes an image having different brightness each been also line divided image an image. Further, the classification unit of the image processing unit classifies each of the line division images having different luminances processed by the line division processing unit into two or more line division image groups according to a predetermined rule, and the integration unit of the image processing unit Thus, for each of the line-divided images classified into the same line-divided image group, the pixel values of the image data obtained by imaging the same position in the inspection target are integrated, and image integrated data is generated for each line-divided image group, The image output unit of the image processing unit outputs the image integration data generated by the integration unit for each line-divided image group according to a predetermined rule. Even if the integrated data is divided and output according to a predetermined rule, the image integration data is divided for each division according to a predetermined rule. Since the appearance of the data of the image is different, easily recognize the type of the inspected defects in image integration data.

  In this case, the two-dimensional image captured at discrete times by the imaging unit has a bright part, a dark part, and a boundary part between the bright part and the dark part, and the transport part inspects the imaging part. It is preferable that the object is relatively transported in the transport direction intersecting the bright part, the dark part, and the boundary part.

  According to this configuration, the two-dimensional image captured at discrete time by the imaging unit has a bright part, a dark part, and a boundary part between the bright part and the dark part. On the other hand, since the inspection object is relatively conveyed in the conveyance direction intersecting the bright part, the dark part, and the boundary part, each part to be inspected in a series of two-dimensional images captured at discrete times is the bright part, the dark part, and the boundary. Enter one of the departments. For this reason, even if the image data output by the division output unit is image integration data in which the pixel values of the image data are integrated, how the image of the image integration data for each division according to a predetermined rule is viewed Therefore, it becomes easier to recognize the type of defect to be inspected in the integrated image data.

  In this case, the classification unit includes a line-divided image group of line-divided images in which the line-divided images processed by the line-division processing unit are lined up in time-series order with bright lines in each of the two-dimensional images. Classification into line-divided image groups of line-divided images in which dark lines in each image are arranged in time-series order and line-divided image groups in line-divided images in which boundary-line lines in each of two-dimensional images are arranged in time-series order It is preferable to do.

  According to this configuration, the line-divided image group of the line-divided images in which each of the line-divided images processed by the line-dividing processing unit by the classification unit is obtained by arranging the bright lines in each of the two-dimensional images in chronological order. A line-divided image group of line-divided images in which dark-line lines in each two-dimensional image are arranged in chronological order, and a line-divided image in line-divided images in which boundary-line lines in each two-dimensional image are arranged in chronological order. Therefore, even if the image data output by the divided output unit is image integrated data in which the pixel values of the image data are integrated, for each line divided image group of the bright part, the dark part, and the boundary part. By dividing and outputting, the image appearance of the image integration data differs for each group of line division images, so the image integration data It becomes more easily recognize the type of defect of the definitive inspected.

  In addition, a light part and a dark part are included in a two-dimensional image captured at discrete times by the imaging unit by shielding a part of light irradiated from the light source to the inspection target between the light source and the inspection target. A light shielding body that forms a boundary portion may be further included, and the conveyance portion may convey the inspection target relative to the light source, the light shielding body, and the imaging unit in a conveyance direction that intersects the bright portion, the dark portion, and the boundary portion.

  According to this configuration, it is possible to easily form a bright part, a dark part, and a boundary part in a two-dimensional image by the light shield, and each part to be inspected in a series of two-dimensional images taken every discrete time is bright. It is possible to enter any one of a part, a dark part, and a boundary part.

  The division output unit is a pixel of the image integration data obtained by imaging the same position in the inspection target, and the luminance in the bright part, the dark part, and the boundary part, of the image integration data for each line division image group generated by the integration part. Are pixels that are equal to or greater than an arbitrary light threshold, pixels of image integration data obtained by imaging the same position in the inspection object, and pixels in which the luminance in the bright part, the dark part, and the boundary part is less than the arbitrary dark threshold, and the same in the inspection object It is preferable to divide and output each of the pixels of the image integration data obtained by imaging the position and having the luminance change width in the bright part, the dark part, and the boundary part equal to or larger than an arbitrary change threshold value.

  According to this configuration, the image integration data for each line division image group generated by the integration unit by the division output unit is a pixel of the image integration data obtained by imaging the same position in the inspection target, and the bright part, the dark part, and the boundary Always bright pixels whose brightness in the area is equal to or higher than the arbitrary light threshold, pixels of the image integration data obtained by imaging the same position in the inspection object, and the luminance in the bright area, dark area and boundary area are lower than the arbitrary dark threshold Always dark pixels and pixels of image integration data obtained by imaging the same position in the inspection object, and pixels with abundant changes in brightness in which the width of the luminance change in the bright part, dark part and boundary part is equal to or greater than an arbitrary change threshold Since it is divided and output separately, it becomes easier to grasp the image of the image integrated data in the bright part, dark part and boundary part for each line divided image group. Easily further recognize the type of defect elephant.

  In this case, the division output unit is a pixel of the image integration data obtained by imaging the same position in the inspection target, and the image integration data for each line division image group generated by the integration unit in the bright part, the dark part, and the boundary part. Pixels whose luminance is greater than or equal to an arbitrary light threshold, pixels of image integration data obtained by imaging the same position in the inspection object, and pixels whose luminance in the bright part, dark part and boundary part are equal to or lower than the arbitrary dark threshold and inspection object Divide each pixel of the image integration data imaged at the same position and whose brightness change width in the bright part, dark part, and boundary part is equal to or greater than an arbitrary change threshold as a colored part colored in different colors. Are preferably output.

  According to this configuration, the image integration data for each line division image group generated by the integration unit by the division output unit is a pixel of the image integration data obtained by imaging the same position in the inspection target, and the bright part, the dark part, and the boundary Always bright pixels whose brightness in the area is equal to or higher than the arbitrary light threshold, pixels of the image integration data obtained by imaging the same position in the inspection object, and the luminance in the bright area, dark area and boundary area are lower than the arbitrary dark threshold Always dark pixels and pixels of image integration data obtained by imaging the same position in the inspection object, and pixels with abundant changes in brightness in which the width of the luminance change in the bright part, dark part and boundary part is equal to or greater than an arbitrary change threshold Each is divided and output as a colored portion colored in a different color. This makes it easy to recognize the appearance of the image of the integrated image data in the bright, dark, and boundary portions of each line-divided image group by the color, so that it is easier to recognize the type of defect to be inspected in the integrated image data. Become.

  In addition, it is preferable that the division output unit divides and outputs the image integration data for each line division image group generated by the integration unit for each line division image group.

  According to this configuration, even in the case of the image integration data in which the pixel values of the image data are integrated, the image of the image integration data for each line division image group is output by being divided for each line division image group. Therefore, it becomes easy to recognize the type of defect to be inspected in the integrated image data.

  In addition, it is preferable that the division output unit divides and outputs the image integration data for each line division image group generated by the integration unit for each predetermined threshold between the line division image groups.

  According to this configuration, even if the image integrated data is obtained by integrating the pixel values of the image data, the line divided image group is output by being divided for each predetermined threshold between the line divided image groups. Since the appearance of the image of the image integration data is different for each of the predetermined threshold values, the type of defect to be inspected in the image integration data can be easily recognized.

  In addition, for each of the line-divided images classified into the same line-divided image group, the integration unit assigns positive / negative signs to the pixels according to the level of the luminance reference value of the pixel of the image data obtained by imaging the same position in the inspection target. It is preferable that the difference values of the image data obtained by imaging the same position in the inspection target are integrated to generate image integration data for each line-divided image group.

  According to this configuration, the integration unit assigns pixels to each of the line division images classified into the same line division image group according to the level of the luminance reference value of the pixel of the image data obtained by imaging the same position in the inspection target. By adding difference values having positive and negative signs, integrating the difference values of the pixels of the image data obtained by imaging the same position in the inspection object, and generating image integrated data for each line-divided image group, it is possible to perform simple calculations. For each line-divided image group, it is possible to generate image integration data in which the height of the pixel data of the image data obtained at the same position in the inspection target is emphasized with respect to the reference value of the luminance.

  In addition, the integration unit performs an enhancement process that emphasizes luminance changes of adjacent pixels of image data obtained by imaging the same position in the inspection target for each of the line division images classified into the same line division image group. It is preferable to integrate the values that have been subjected to pixel enhancement processing of image data obtained by imaging the same position in to generate image integrated data for each line-divided image group.

  According to this configuration, the integration unit performs an enhancement process for emphasizing luminance changes of pixels adjacent to each other in the image data obtained by imaging the same position in the inspection target for each of the line divided images classified into the same line divided image group. In order to generate integrated image data for each group of line-divided images by integrating values that have been subjected to pixel enhancement processing of image data obtained by imaging the same position in the inspection object, a line-divided image can be obtained by a simple calculation. For each group, it is possible to generate image integration data in which luminance changes of pixels adjacent to each other in image data obtained by imaging the same position in the inspection target are emphasized.

  And an analysis unit for identifying the type of defect to be inspected based on data accumulated as a result of machine learning related to identification of the type of defect included in the integrated image data for each line-divided image group generated by the integration unit. It is suitable to provide.

  According to this configuration, the analysis unit determines the defect to be inspected based on the data accumulated by the result of machine learning related to the identification of the type of defect included in the image integration data for each line-divided image group generated by the integration unit. Although the type is identified, the image integration data obtained by integrating the pixel values of the image data is integrated with the pixel values of the image data for each line-divided image group. Since the defect type is identified based on the result of the above, the identification accuracy of the defect type to be inspected can be improved.

  On the other hand, the present invention provides an irradiation process for irradiating the inspection object with light from the light source of the defect inspection system, and light that has been transmitted from the light source to the inspection object through the irradiation process and transmitted or reflected by the imaging unit of the defect inspection system. An imaging process for capturing a two-dimensional image by discrete time, a transport process for transporting an inspection object relative to the light source and the imaging section in the transport direction by a transport section of the defect inspection system, and an image of the defect inspection system An image processing step for processing the image data of the two-dimensional image picked up in the image pickup step by the processing unit. In the image pickup step, a two-dimensional image whose luminance changes in a direction matching the transport direction in the two-dimensional image is picked up In the image processing step, the two-dimensional image is divided into a plurality of lines arranged in parallel in the conveyance direction, and each of the two-dimensional images captured at discrete times by the imaging unit is used. A line division processing step for processing the image data of a line division image in which the lines at the same position are arranged in chronological order, and two or more line division images according to a predetermined rule for each of the line division images processed in the line division step For each of the classification step for classifying into a group and the line division images classified into the same line division image group in the classification step, the pixel values of the image data obtained by imaging the same position in the inspection object are integrated, and the line division image group This is a defect inspection method having an integration step of generating image integration data every time and a division output step of dividing and outputting the image integration data for each line division image group generated in the integration step according to a predetermined rule.

  In this case, in the irradiation process, the two-dimensional image captured at discrete time in the imaging process has a bright part, a dark part, and a boundary part between the bright part and the dark part. It is preferable that the inspection object is relatively transported in the transport direction intersecting the bright part, the dark part, and the boundary part.

  In this case, in the classification step, each of the line division images processed in the line division processing step is a group of line division images of a line division image in which bright lines in each of the two-dimensional images are arranged in chronological order. Classification into line-divided image groups of line-divided images in which dark lines in each image are arranged in time-series order and line-divided image groups in line-divided images in which boundary-line lines in each of two-dimensional images are arranged in time-series order It is preferable to do.

  In the irradiation process, a two-dimensional image captured every discrete time in the imaging process is provided by a light shielding body that is located between the light source and the inspection target and shields a part of the light irradiated from the light source to the inspection target. It is preferable that the bright part, the dark part, and the boundary part are formed, and in the transporting process, the inspection target is transported relative to the light source, the light shielding body, and the imaging unit in the transport direction intersecting the bright part, the dark part, and the boundary part. .

  Further, in the divided output step, the image integration data for each line division image group generated in the integration step is a pixel of the image integration data obtained by imaging the same position in the inspection target, and the luminance in the bright part, the dark part, and the boundary part Are pixels that are equal to or greater than an arbitrary light threshold, pixels of image integration data obtained by imaging the same position in the inspection object, and pixels in which the luminance in the bright part, the dark part, and the boundary part is less than the arbitrary dark threshold, and the same in the inspection object It is preferable to divide and output each of the pixels of the image integration data obtained by imaging the position and having the luminance change width in the bright part, the dark part, and the boundary part equal to or larger than an arbitrary change threshold value.

  In this case, in the division output step, the image integration data for each line division image group generated in the integration step is a pixel of the image integration data obtained by imaging the same position in the inspection target, and in the bright portion, the dark portion, and the boundary portion. Pixels whose luminance is greater than or equal to an arbitrary light threshold, pixels of image integration data obtained by imaging the same position in the inspection object, and pixels whose luminance in the bright part, dark part and boundary part are equal to or lower than the arbitrary dark threshold and inspection object Divide each pixel of the image integration data imaged at the same position and whose brightness change width in the bright part, dark part, and boundary part is equal to or greater than an arbitrary change threshold as a colored part colored in different colors. Are preferably output.

  In the divided output step, it is preferable that the image integration data for each line division image group generated in the integration step is divided and output for each line division image group.

  In the divided output step, it is preferable to divide and output the image integration data for each line divided image group generated in the integrating step for each predetermined threshold between the line divided image groups.

  Further, in the integration step, for each of the line division images classified into the same line division image group, a positive or negative sign is assigned to the pixel according to the level of the luminance reference value of the pixel of the image data obtained by imaging the same position in the inspection target. It is preferable that the difference values of the image data obtained by imaging the same position in the inspection target are integrated to generate image integration data for each line-divided image group.

  Further, in the integration step, for each of the line-divided images classified into the same line-divided image group, an emphasis process for emphasizing the luminance change of adjacent pixels of the image data obtained by imaging the same position in the inspection object is performed, and the inspection object It is preferable to integrate the values that have been subjected to pixel enhancement processing of image data obtained by imaging the same position in to generate image integrated data for each line-divided image group.

  In addition, an analysis step of identifying the type of defect to be inspected based on data accumulated as a result of machine learning related to identification of the type of defect included in the integrated image data for each line-divided image group generated in the integration step It is suitable to provide.

  According to the defect inspection system and the defect inspection method of the present invention, it becomes easy to recognize the type of defect.

It is a perspective view which shows the defect inspection system which concerns on embodiment. It is a figure which shows arrangement | positioning of the light source of the defect inspection system of FIG. 1, an imaging part, a light shielding body, and a test object. It is a block diagram which shows the detail of the image process part of the defect inspection system of FIG. It is a flowchart which shows the process of the defect inspection method which concerns on embodiment. 5 is a flowchart showing details of the image processing step in FIG. 4. (A), (B), (C), (D), (E), (F), and (G) are images processed by the line division processing unit of the image processing unit of the defect inspection system of FIG. FIG. (A) is a figure which shows a time-series two-dimensional image, (B) is a figure which shows each of the line division | segmentation image which arranged the line of each position in time series order, (C) is a figure of (B). It is a figure which shows the alignment image which shifted time so that each of a line division | segmentation image might show the same position of a test object. (A), (B), and (C) are diagrams showing images processed by the classification unit of the image processing unit of the defect inspection system of FIG. 1, and (D), (E), and (F) are It is a figure which shows the image processed by the integration part of the image processing part of the defect inspection system of FIG. FIG. 9 is a diagram illustrating a state in which each pixel of the image integration data in FIG. 8D is output as a colored portion that is different from each other by the divided output unit of the defect inspection system in FIG. 1. It is a figure which shows a convolution neural network.

  Hereinafter, preferred embodiments of a defect inspection system and a defect inspection method of the present invention will be described in detail with reference to the drawings. As shown in FIGS. 1 and 2, the defect inspection system 1 according to the embodiment of the present invention includes a light source 2, an imaging unit 3, a transport unit 4, an image processing unit 5, a light shield 6, a parallel light lens 7, and an analysis unit. 8 is provided. The defect inspection system according to the present embodiment detects defects of the inspection target T using an optical film such as a polarizing film and a retardation film, and a film such as a laminated film used for a battery separator as the inspection target T. The inspection target T extends in the transport direction X of the transport unit 4 and has a preset width in the width direction Y orthogonal to the transport direction X. The defect generated in the inspection target T refers to a state different from a desired state. For example, a foreign object, a dent, a bubble (such as an object generated at the time of molding), a foreign object bubble (an object generated due to mixing of an object), Examples include scratches, nicks (such as those caused by crease marks), and streaks (such as those caused by differences in thickness). The defect inspection system 1 identifies these defect types.

  As shown in FIGS. 1 and 2, the light source 2 irradiates the inspection target T with light. The light source 2 is arranged to irradiate linear light parallel to the width direction Y. The light source 2 is not particularly limited as long as it irradiates light that does not affect the composition and properties of the film to be inspected T, such as a metal halide lamp, a halogen transmission light, and a fluorescent lamp.

  The imaging unit 3 captures a two-dimensional image of light that is irradiated from the light source 2 onto the inspection target T and transmitted or reflected by the inspection target T at discrete time intervals. The imaging unit 3 includes a plurality of optical members and photoelectric conversion elements. The optical member is composed of an optical lens, a shutter, and the like, and forms an image on the surface of the photoelectric conversion element of light that has passed through the film that is the inspection target T. The photoelectric conversion element is an area sensor composed of an imaging element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal-Oxide Semiconductor) that captures a two-dimensional image. The imaging unit 3 may capture either a two-dimensional image having no color or a two-dimensional image having a color.

  The transport unit 4 transports the inspection target T relative to the light source 2 and the imaging unit 3 in the transport direction X. The transport unit 4 includes, for example, a sending roller and a receiving roller for transporting the film to be inspected T in the transport direction X, and measures the transport distance using a rotary encoder or the like. In this embodiment, the conveyance speed of the inspection target T by the conveyance unit 4 is set to about 2 to 100 m / min in the conveyance direction X. The conveyance speed in the conveyance unit 4 is set and controlled by the image processing unit 5 and the analysis unit 8.

  The image processing unit 5 processes the image data of the two-dimensional image captured by the imaging unit 3. The image processing unit 5 is not particularly limited as long as it can perform image processing of two-dimensional image data. For example, a PC (personal computer) in which image processing software is installed and an image processing circuit are described. An image capture board or the like equipped with an FPGA (Field Programmable Gate Array) can be applied.

  The light-shielding body 6 is located between the light source 2 and the inspection target T, and is imaged 2 every discrete time by the imaging unit 3 by shielding part of the light emitted from the light source 2 to the inspection target T. A bright part, a dark part, and a boundary part between the bright part and the dark part are formed in the three-dimensional image. With the light shield 6, the imaging unit 3 captures a two-dimensional image whose luminance changes in a direction that matches the conveyance direction X in the two-dimensional image. More specifically, the transport unit 4 is relative to the light source 2, the parallel light lens 7, the light shield 6, and the imaging unit 3 in the transport direction X where the inspection target T intersects the bright part, the dark part, and the boundary part. Transport to. In the present embodiment, the longitudinal direction of the boundary portion is parallel to the width direction Y perpendicular to the transport direction X. Note that the light-shielding body 6 may not be provided as long as the imaging unit 3 can capture a two-dimensional image whose luminance changes in a direction that matches the conveyance direction X in the two-dimensional image. The parallel light lens 7 makes the traveling direction of light irradiated from the light source 2 to the inspection target T and the light shielding body 6 parallel. The parallel light lens 7 can be configured by a telecentric optical system, for example.

  Details of the image processing unit 5 will be described below. As shown in FIG. 3, the image processing unit 5 includes a line division processing unit 9, a classification unit 10, an integration unit 11, and a division output unit 12. The line division processing unit 9 divides the two-dimensional image into a plurality of lines arranged in parallel in the transport direction X, and parallels the lines at the same position in each of the two-dimensional images picked up by the image pickup unit 3 at discrete times in order of time series. Process the image data of the divided line image.

  The classification unit 10 classifies each of the line division images processed by the line division processing unit 9 into two or more line division image groups according to a predetermined rule. The integration unit 11 integrates pixel values of image data obtained by imaging the same position in the inspection target T for each of the line division images classified into the same line division image group, and the image integration data for each line division image group. Is generated.

  More specifically, the integration unit 11 performs line division of the line division image in which each of the line division images processed by the line division processing unit 9 is arranged in parallel with the bright line in each of the two-dimensional images in time series. Line grouped image group of line segmented images in which dark line lines in each of the image group and two-dimensional image are arranged in chronological order, and line segmented image lines in which boundary line lines in each of the two-dimensional images are arranged in chronological order Classify into divided image groups.

  Further, the integration unit 11 determines whether the pixels are positive or negative according to the level of the pixel luminance reference value of the image data obtained by imaging the same position in the inspection target T for each of the line divided images classified into the same line divided image group. A difference value having a sign is assigned, and pixel difference values of image data obtained by imaging the same position in the inspection object are integrated to generate image integration data for each line-divided image group. Alternatively, the integration unit 11 performs an enhancement process that emphasizes the luminance change of pixels adjacent to each other in the image data obtained by imaging the same position in the inspection target for each of the line division images classified into the same line division image group, The values obtained by emphasizing the pixels of the image data obtained by imaging the same position in the object are integrated to generate image integrated data for each line divided image group.

  The division output unit 12 divides and outputs the image integration data for each line division image group generated by the integration unit 11 according to a predetermined rule. More specifically, the division output unit 12 divides and outputs the image integration data for each line division image group generated by the integration unit 11 for each line division image group. Or a division | segmentation output part divides | segments and outputs the image integration data for every line division | segmentation image group produced | generated by the integration part for every predetermined threshold between line division | segmentation image groups. The predetermined threshold is, for example, the luminance of the pixels of the line divided image belonging to the line divided image group.

  The division output unit 12 is a pixel of image integration data obtained by dividing the image integration data for each line division image group generated by the integration unit 11 for each line division image group and imaging the same position in the inspection target T. Pixels whose brightness in the bright part, dark part, and boundary part is greater than or equal to an arbitrary bright threshold, and pixels of image integration data obtained by imaging the same position in the inspection target T, and the brightness in the bright part, dark part, and boundary part is arbitrary. Pixels that are equal to or less than the dark threshold value, and pixels of image integration data obtained by imaging the same position on the inspection target T, and each of the pixels whose brightness change widths in the bright part, the dark part, and the boundary part are equal to or greater than an arbitrary change threshold value Divided into and output.

  Further, the divided output unit 12 is a pixel of image integrated data obtained by imaging the same position in the inspection target T, which is the image integrated data for each line divided image group generated by the integrating unit 11, and includes a bright part, a dark part, and a boundary part. Pixels whose luminance at or above the arbitrary light threshold is equal to, pixels of the image integration data obtained by imaging the same position on the inspection object, and pixels whose luminance at the bright part, dark part and boundary part are below the arbitrary dark threshold and inspection As a colored portion in which the pixels of the image integration data obtained by imaging the same position in the object and each pixel whose brightness change width in the bright portion, the dark portion, and the boundary portion is equal to or greater than an arbitrary change threshold are colored differently from each other Divide and output.

  Returning to FIG. 2, the analysis unit 8 connected to the image processing unit 5 includes, for example, a PC (personal computer). The analysis unit 8 identifies the defect type of the inspection target T based on the data accumulated as a result of machine learning related to the identification of the type of defect included in the image integration data for each line-divided image group generated by the integration unit 11. To do. Data obtained by accumulating the results of machine learning is stored in a storage device such as a hard disk of a PC including the analysis unit 8 and is updated according to the results of machine learning.

  In the present embodiment, the data obtained by accumulating the results of machine learning related to the identification of the type of defect included in the line-divided image is a series of images captured at discrete times by the imaging unit 3 in the defect inspection system 1. In addition to the data storing the results of machine learning related to the identification of the type of defect included in the line segmented image obtained by processing the two-dimensional image, the defect included in the line segmented image separately created outside the defect inspection system 1 Data that stores the results of machine learning related to type identification is also included. That is, in this embodiment, in addition to the aspect in which the type of defect is identified in a state in which machine learning is performed in the defect inspection system 1, the defect is in a state in which machine learning has not yet been performed in the defect inspection system 1. A mode is also included in which the type of defect is identified based on the data stored in the machine learning result separately created outside the inspection system 1.

  The analysis unit 8 also displays the type of defect identified by the image processing unit 5 on an LC (Liquid Crystal) display panel, a plasma display panel, an EL (ElectroLuminescence) display panel, or the like. The image processing unit 5 may have an analysis unit 8 that identifies the type of defect and displays the processed image.

  Hereinafter, the defect inspection method of this embodiment will be described. As shown in FIG. 4, the irradiation process of irradiating the inspection target T with light from the light source 2 of the defect inspection system 1 is performed (S1). As shown in FIG. 6A, in the irradiation process, the defect inspection system 1 is located between the light source 2 and the inspection target T and blocks part of the light irradiated from the light source 2 to the inspection target T. The body 6 forms a bright part l, a dark part d, and a boundary part b between the bright part l and the dark part d in the two-dimensional image F (t1) picked up at discrete times in the imaging step. . As shown in FIG. 6A, the two-dimensional image F (t1) at time t = t1 is two-dimensional as it goes downstream in the transport direction X because the light from the light source 2 is shielded by the light shield 6. The brightness in the image F (t1) increases. Further, the defect D on the film of the inspection target T is shown in the two-dimensional image F (t1). The same applies to the two-dimensional images F (t2), F (t3),..., F (tm) at time t = t2, t3, ..., tm (m is an arbitrary natural number).

  As shown in FIG. 4, the imaging unit 3 of the defect inspection system 1 generates a two-dimensional image F (t1) by light irradiated from the light source 2 to the inspection target T through the irradiation process and transmitted or reflected by the irradiation process in discrete time. An imaging step of imaging every time is performed (S2). As shown in FIG. 6A, in the imaging step, a part of light irradiated from the light source 2 to the inspection target T is shielded by the light shield 6, so that the conveyance direction X in the two-dimensional image F (t1) A two-dimensional image F (t1) whose luminance changes in the matching direction is captured. The same applies to the two-dimensional images F (t2), F (t3),..., F (tm) at time t = t2, t3,.

  Further, as shown in FIG. 4, a transport step of transporting the inspection target T relative to the light source 2 and the imaging unit 3 in the transport direction X is performed by the transport unit 4 of the defect inspection system 1 (S <b> 3). As shown in FIG. 6A, in the transporting process, the inspection target T is transported across the light part 1, the dark part d, and the boundary part b with respect to the light source 2, the parallel light lens 7, the light shield 6 and the imaging unit 3. Transport relative to direction X. In the present embodiment, the boundary b is parallel to the width direction Y perpendicular to the transport direction X, but the angle formed by the boundary b and the transport direction X may be other than 90 °. Further, the boundary portion b is not necessarily strict, and the boundary portion is the portion with the highest luminance of the two-dimensional image F (t1) included in the bright portion l and the luminance of the two-dimensional image F included in the dark portion d. This means a site intermediate to the site.

  As shown in FIG. 4, the image processing unit 5 of the defect inspection system 1 performs an image processing process for processing the image data of the two-dimensional images F (t1) to F (tm) captured in the imaging process (S4). ). Details of the image processing process will be described below. As shown in FIG. 5, in the image processing step, the line division processing step is performed by the line division processing unit 9 of the image processing unit 5 of the defect inspection system 1 (S41). As shown in FIG. 6B, in the line division processing step, the line division processing unit 9 includes a plurality of first lines L1 (t1) to jth in which the two-dimensional image F (t1) is aligned in the transport direction X. Line Lj (t1) to k-th line Lk (t1) (j and k are arbitrary natural numbers, j ≦ k). The width of the line L1 (t1) to the line Lk (t1) in the transport direction X is such that the inspection target T is in the transport direction X at each frame interval of time t1, time t2,..., Time tj,. It is the same as the distance conveyed. Similar processing is performed for the two-dimensional images F (t2), F (t3),..., F (tm) at time t = t2, t3,.

  The line division processing unit 9 is a line L1 at the same position in each of the two-dimensional images F (t1) to F (tm) obtained by capturing the two-dimensional images F (t1) to F (tm) at discrete times in the imaging process. (T1), L1 (t2), L1 (t3), etc. are processed into image data of line-divided images in which time series are arranged in parallel. The first line division image will be described as an example. As shown in FIG. 6 (C), the line division processing unit 9 performs the conveyance direction X in each of the two-dimensional images F (t1), F (t2), F (t3),. The most downstream first lines L1 (t1), L1 (t2), L1 (t3),... Are arranged in parallel in time series (conveying direction X). As shown in FIG. 6D, the line division processing unit 9 parallels the first lines L1 (t1) to L1 (tm) in the two-dimensional images F (t1) to F (tm) in chronological order. Thus, the first line divided image DL1 (t1) is generated.

  As shown in FIGS. 6E, 6F, and 6G, the line division processing unit 9 includes the first line L1 (2) in each of the two-dimensional images F (t1) to F (tm). t1) to L1 (tm),..., j-th line Lj (t1) to Lj (tm),..., k-th line Lk (t1) to Lk (tm) Line division image DL1 (t1), ..., j-th line division image DLJ (t1), ..., k-th line division image DLk (t1) are generated. As shown in FIG. 6E, the line-divided image DL1 (t1) represents the lines L1 (t1) to L1 (tk) at the positions of the bright portions 1 in the two-dimensional images F (t1) to F (tk). They are arranged in series order.

  As shown in FIGS. 7A and 7B, the line-divided images DL1 (t1) to DLk (t1) are two-dimensional images F (t1) to F (tm) captured every discrete time. Since the lines L1 (t1) to Lk (t1) at the same position in each are arranged in chronological order, the line divided images DL1 (t1) to DLk (t1) in the same time range are the inspection target T. The positions of the defects D in the line-divided images DL1 (t1) to DLk (t1) are also shifted from each other. Therefore, in the present embodiment, each line-divided image is inspected by creating a line-divided image in which lines at the same position in two-dimensional images captured in different time ranges are arranged in chronological order. Alignment is performed to indicate the same position of T.

  As shown in FIG. 7A, two-dimensional images F (t1) to F (tm) are imaged at discrete times in the imaging process. Since the inspection target T is transported in the transport direction X, the positions of the defects D in the two-dimensional images F (t1) to F (tm) are shifted from each other. As shown in FIG. 7B, line division images DL1 (t1) to DLj (t1) to DLk (t1) are generated as described above. Since the line division images DL1 (t1) to DLk (t1) in the same time range indicate different positions of the inspection target T, the position of the defect D in the line division images DL1 (t1) to DLk (t1) is also They are shifted.

  For the first line L1 (t1) to L1 (tm) from the downstream side in the transport direction X, for example, the jth line Lj (t1) to Lj (tm) from the downstream side in the transport direction X in the same time range. ) Indicates the upstream position of the inspection target T in the transport direction X by the distance by which the inspection target T is transported at a frame interval of (j−1). Therefore, as shown in FIG. 7C, for example, the j-th line L1 (tm) to L1 (t (m + (m−1))) of the line-divided image DL1 (tm) With respect to the line-divided image of the line, the time t (m− (j−1)) to the time t (m + (m +) that is back by the time of the frame interval of (j−1) with respect to the range from the time t1 to the time tm. -J)) in the range of the line division image DLj (t (m- (j-1))) indicates the same position of the inspection target T.

  Similarly, for the line division image DL1 (tm) of the first line L1 (tm) to L1 (t (m + (m-1))), for example, for the line division image of the kth line, Line division in the range from time t (m− (k−1)) to time t (m + (m−k)) that is back by the time of the frame interval of (k−1) with respect to the range from t1 to time tm. The image DLk (t (m− (k−1))) indicates the same position of the inspection target T.

  Alternatively, for example, with respect to the line division image DL1 (t1) of the first line L1 (t1) to L1 (t (1+ (m-1))), for the line division image of the jth line, the time t A line division image DLj (t (1- (j-1))) in the range from (1- (j-1)) to time t (1+ (m-j)) indicates the same position of the inspection target T. In addition, for the line division image DL1 (t1) of the first line L1 (t1) to L1 (t (1+ (m-1))), for example, for the line division image of the kth line, the time t A line division image DLk (t (1- (k-1))) in a range from (1- (k-1)) to time t (1+ (m-k)) indicates the same position of the inspection target T. By shifting the time range in this way, it is possible to perform alignment so that each of the line divided images indicates the same position of the inspection target T.

  For example, the line division image DLj (t (1- (j−1))) illustrated in FIG. 6F is a two-dimensional image F (t (1- (j−1))) to F (t (m− (J-1))) in which the lines Lj (t (1- (j-1))) to Lj (t (m- (j-1))) at the position of the boundary b are arranged in parallel in chronological order. It is. In addition, the line division image DL (k−2) (t (1− (k−3))) illustrated in FIG. 6G is a two-dimensional image F (t (1− (k−3))) to F. Lines L (k-2) (t (1- (k-3))) to Lk (t (m- (k-3)) at the position of the dark part d at (t (m- (k-3))) ) In parallel in chronological order.

  If the amount of misalignment is known, or if the size of the line division image is sufficiently large relative to the defect, the defect will always fit in the line division image, so the line division that includes the defect without alignment Images can be used for machine learning. Therefore, in such a case, alignment may not be performed.

  As shown in FIG. 5, in the image processing step, the classification step is performed by the classification unit 10 of the image processing unit 5 of the defect inspection system 1 (S42). In the classification step, the classification unit 10 classifies each of the line division images DL1 (t1) processed in the line division step into two or more line division image groups according to a predetermined rule.

  As shown in FIG. 8A, in the classification step, the classification unit 10 converts each of the line division images DL1 (t1) and the like processed in the line division processing step in each of the two-dimensional images F (t1) and the like. Line L1 (t1), DL2 (t0), DL3 (t) in which lines L1 (t1)..., L2 (t0)..., L3 (t (-1)). (-1)), and so on, are divided into line-divided image groups G1 (t1).

  As shown in FIG. 8B, in the classification step, the classification unit 10 performs two-dimensional processing on each of the line division images DLj (t (1- (j-1))) and the like processed in the line division processing step. Lines Lj (t (1- (j-1)))..., L (j + 1) (t (1-j)) at the boundary b in each of the images F (t (1- (j-1))), etc. .., L (j + 2) (t (1- (j + 1)))... Are line-divided images DLj (t (1- (j−1))), DL (j + 1) (t (1) −j)), DL (j + 2) (t (1− (j + 1)))..., And line segmented image group G2 (t (1− (j−1))).

  As shown in FIG. 8C, in the classification step, the classification unit 10 includes the line division image DL (k-2) (t (1- (k-3))) processed in the line division processing step. Each of the lines L (k−2) (t (1− (k−3))) of the dark part d in each of the two-dimensional images F (t (1− (k−3))) etc., L (k -1) (t (1- (k-2))..., Lk (t (1- (k-1))). 1- (k-3))), DL (k-1) (t (1- (k-2))), DLk ((1- (k-1)))... (1- (k-3))).

  As shown in FIG. 5, in the image processing step, the integration step is performed by the integration unit 11 of the image processing unit 5 of the defect inspection system 1 (S43). As shown in FIG. 8D, in the integration step, the integration unit 11 performs line division images DL1 (t1), DL2 (t0), DL3 (t (−) classified into the same line division image group G1 (t1). For each of 1)),..., The pixel values of the image data obtained by imaging the same position on the inspection target T are integrated to generate image integration data C1 (t1) for each line-divided image group G1 (t1).

  As shown in FIG. 8 (E), in the integration step, the integration unit 11 performs line division images DLj (t (1- (1)) classified into the same line division image group G2 (t (1- (j-1))). (J-1))), DL (j + 1) (t (1-j)), DL (j + 2) (t (1- (j + 1)))... The pixel values of the data are integrated to generate image integration data C2 (t (1- (j-1))) for each line-divided image group G2 (t (1- (j-1))).

  As shown in FIG. 8F, in the integration step, the integration unit 11 performs line division images DL (k-2) classified into the same line division image group G3 (t (1- (k-3))). (T (1- (k-3))), DL (k-1) (t (1- (k-2))), DLk ((1- (k-1)))... The pixel values of the image data obtained by imaging the same position in the target T are integrated, and the image integration data C3 (t (1- (k−)) is obtained for each line divided image group G3 (t (1- (k−3))). 3))) is generated.

  In the integration step, the integration unit 11 performs the processing for each of the line division images DL1 (t1), DL2 (t0), DL3 (t (-1)), etc. classified into the same line division image group G1 (t1). A difference value having a positive / negative sign is given to the pixel according to the level of the luminance reference value of the pixel of the image data captured at the same position on the inspection target T, and the pixel of the image data captured at the same position on the inspection target T The difference values may be integrated to generate image integration data C1 (t1) or the like for each line divided image group G1 (t1) or the like.

  Alternatively, in the integration step, the integration unit 11 includes each of the line division images DL1 (t1), DL2 (t0), DL3 (t (-1)), etc. classified into the same line division image group G1 (t1). A value obtained by performing an emphasis process for emphasizing luminance changes of pixels adjacent to each other in image data obtained by imaging the same position in the inspection target T, and by performing an emphasis process on pixels of image data obtained by imaging the same position in the inspection target T. They may be integrated to generate image integration data C1 (t1) or the like for each line-divided image group G1 (t1) or the like. Note that the examples shown in FIGS. 8D, 8E, and 8F are schematic diagrams only, and the actual image integration data C1 (t1) and the like are displayed according to the shape of the defect D. Depends on etc.

  As shown in FIG. 5, in the image processing step, the division output step is performed by the division output unit 12 of the image processing unit 5 of the defect inspection system 1 (S44). In the division output step, the division output unit 12 divides and outputs the image integration data C1 (t1) and the like for each line division image group G1 (t1) and the like generated in the integration step according to a predetermined rule. For example, in the divided output step, the divided output unit 12 outputs the image integration data C1 (t1) and the like for each line divided image group G1 (t1) generated in the integration step for each line divided image group G1 (t1) and the like. Divide and output. First, in the divided output step, the divided output unit 12 outputs the image integration data C1 (t1) and the like for each line divided image group G1 (t1) generated in the integration step for each line divided image group G1 (t1) and the like. To divide.

  As shown in FIG. 9, in the divided output step, the divided output unit 12 is a pixel such as image integration data C1 (t1) obtained by imaging the same position in the inspection target T, and includes a bright portion l, a dark portion d, and a boundary portion b. Are pixels that are equal to or larger than an arbitrary bright threshold, and pixels such as image integration data C1 (t1) obtained by imaging the same position on the inspection target T, and the luminance at the bright portion l, the dark portion d, and the boundary portion b is arbitrary. Pixels that are equal to or less than the dark threshold value, and pixels such as image integration data C1 (t1) obtained by imaging the same position in the inspection target T, and the widths of changes in luminance at the bright portion l, the dark portion d, and the boundary portion b are arbitrary. Each pixel that is equal to or greater than the change threshold is output as a colored portion R colored red, a colored portion B colored blue, and a colored portion G colored green.

  For example, the divided output unit 12 is a pixel such as the image integration data C1 (t1) obtained by imaging the same position in the inspection target T, and the luminance in the bright part l, the dark part d, and the boundary part b is always equal to or higher than the bright threshold. Each bright pixel is assigned to the colored portion R, and is a pixel such as image integration data C1 (t1) obtained by imaging the same position in the inspection target T, and the luminance in the bright portion l, the dark portion d, and the boundary portion b is equal to or less than the dark threshold value. Each of the always dark pixels is assigned to the colored portion B, and is a pixel such as the image integration data C1 (t1) obtained by imaging the same position on the inspection target T, and the luminance at the bright portion l, the dark portion d, and the boundary portion b. Each of the pixels rich in changes in brightness whose change width is equal to or greater than the change threshold value can be assigned to the colored portion G.

  The coloring part R, the coloring part B, and the coloring part G may be assigned in any manner. In addition, the light threshold value, the dark threshold value, and the change threshold value can be set in any way, but generally, the light threshold value ≧ the dark threshold value.

  For example, in the divided output process, the divided output unit 12 converts the image integrated data C1 (t1) and the like generated for the line divided image group G1 (t1) and the like generated in the integrating process into the line divided image group G1 (t1) and the like. You may divide and output for every predetermined threshold value.

  As shown in FIG. 4, the analysis unit 8 of the defect inspection system 1 identifies the type of defect included in the image integration data C1 (t1) etc. for each of the line divided image groups G1 (t1) etc. generated in the integration process. An analysis step for identifying the type of the defect D of the inspection target T based on the data obtained by accumulating the machine learning result is performed (S5). For machine learning, for example, a convolutional neural network is performed. Note that a neural network other than the convolutional neural network and other methods can be used as long as the type of defect can be identified by machine learning.

  As shown in FIG. 10, the convolutional neural network 100 includes an input layer 110, a hidden layer 120, and an output layer 130. In the input layer 110, the line-divided image groups G1 (t1), G2 (t (1- (j-1))), G3 (t (t ( 1- (k-3))) and the like, image integration data C1 (t1), C2 (t (1- (j-1))), C3 (t (1- (k-3))) are input. The The hidden layer 120 includes convolution layers 121 and 123 on which image processing by the weighting filter f is performed, and a pooling layer 122 that performs processing for reducing the two-dimensional array output from the convolution layers 121 and 123 in the vertical and horizontal directions and leaving an effective value. And a total coupling layer 124 in which the weight coefficient n of each layer is updated. In the output layer 130, the identification result of the type of the defect D by machine learning is output. In the convolutional neural network 100, the weight of each layer is learned by back-propagating the error between the output identification result and the correct answer value in the reverse direction R.

  For example, by inputting a plurality of pieces of image integration data C1 (t1) together with correct answers for identifying the type of defect D to the analysis unit 8 in advance, the newly input image integration data C1 (t1) and the like are input. It is sequentially identified whether the contained object is the type of a specific defect D, and the identification result is sequentially output. The error between the identification result and the correct answer sequentially output is back-propagated in the reverse direction R, and the weight coefficient n of each layer is sequentially updated and stored as data. With the weight of each phase sequentially updated, whether or not the object included in the newly input image integration data C1 (t1) is a specific defect type is sequentially identified, and the identification results are sequentially output. The weight coefficient n of each layer is sequentially updated based on the error between the sequentially output identification result and the correct answer, and is repeatedly stored as data, thereby reducing the error between the identification result and the correct answer, The accuracy of identifying the type of D is improved.

  In the present embodiment, the light source 2 that irradiates the inspection target T with light and the two-dimensional image F (t1) that is emitted from the light source 2 to the inspection target T and transmitted or reflected by the inspection target T are captured at discrete time intervals. Image data of the two-dimensional image F (t1) imaged by the imaging unit 3, the conveyance unit 4 that conveys the inspection target T relative to the light source 2 and the imaging unit 3 in the conveyance direction X In the defect inspection system 1 including the image processing unit 5 that processes the two-dimensional image F (t1) whose luminance changes in the direction matching the conveyance direction X in the two-dimensional image F (t1) by the imaging unit 3. The two-dimensional image F (t1) is divided into a plurality of lines L1 (t1) arranged in parallel in the transport direction X by the line division processing unit 9 of the image processing unit 5 and is imaged by the imaging unit 3 at discrete times. Of the two-dimensional image F (t1) Since the image data such as the line division image DL1 (t1) in which the lines L1 (t1) and the like at the same position in each are arranged in chronological order are processed, even if the images of the same inspection target T are captured Each of the divided images DL1 (t1) and the like is an image having different luminance.

  In addition, each of the divided line images DL1 (t1) having different luminances processed by the line dividing processing unit 9 by the classification unit 10 of the image processing unit 5 is divided into two or more line divided image groups G1 ( t1) and the like, and the integration unit 11 of the image processing unit 5 sets the same position in the inspection target T for each of the line division images DL1 (t1) and the like classified into the same line division image group G1 (t1) and the like. Pixel values of the captured image data are integrated to generate image integration data C1 (t1) for each line division image group G1 (t1) and the like, and the integration unit 11 is generated by the division output unit 12 of the image processing unit 5. Since the image integrated data C1 (t1) etc. for each line divided image group G1 (t1) etc. generated by the above are divided and output according to a predetermined rule, the image data Even in the case of image integration data C1 (t1) or the like in which prime values are integrated, the image integration data C1 (t1) is divided for each division according to a predetermined rule by being divided and output according to a predetermined rule. ) And the like are different from each other, so that it becomes easy to recognize the type of the defect D of the inspection target T in the image integration data C1 (t1) or the like.

  Further, according to the present embodiment, the two-dimensional image F (t1) captured at discrete time by the imaging unit 3 is a bright part l, a dark part d, and a boundary part between the bright part l and the dark part d. b, and the inspection unit is conveyed relative to the imaging unit 3 in the conveyance direction X that intersects the bright part l, the dark part d, and the boundary part b, and thus is imaged at discrete times. Each part of the inspection target T in the series of two-dimensional images F (t1)... Enters one of the bright part l, the dark part d, and the boundary part b. For this reason, even if the image data output by the division output unit 12 is image integration data in which the pixel values of the image data are integrated, the appearance of the image of the image integration data for each division according to a predetermined rule Since the direction is further different, it becomes easier to recognize the type of the defect D of the inspection target T in the image integration data.

  In addition, according to the present embodiment, each of the line-divided images processed by the line-dividing processor 9 by the classifier 10 applies the lines of the bright part l in each of the two-dimensional images F (t1) and the like in chronological order. Line division image DL (k−2) in which the lines of the dark part d in the line division image group G1 (t1), the two-dimensional image F (t1), etc. of the line division image DL1 (t1) arranged in parallel are arranged in time series. ) (T (1- (k-3))) line division image group G3 (t (1- (k-3))), two-dimensional image F (t1), etc. Since it is classified into the line-divided image group G2 (t (1- (j-1))) of the line-divided images DLj (t (1- (j-1))) arranged in series, the divided output unit 12 The output image data is the pixel of the image data. Even if the integrated image data is integrated with each other, the divided line image group G1 (t1) is output by being divided for each line divided image group G1 (t1) of the bright part 1, the dark part d, and the boundary part b. ) And the like, the appearance of the image of the image integration data is further different, so that it becomes easier to recognize the type of the defect D of the inspection target T in the image integration data.

  In addition, according to the present embodiment, the light portion l, the dark portion d, and the boundary portion b can be easily formed in the two-dimensional image F (t1) or the like by the light shield 6, and a series of images taken at discrete times. In the two-dimensional image F (t1)..., Each part of the inspection target T can enter any one of the bright part l, the dark part d, and the boundary part b.

  In addition, according to the present embodiment, the integrated image data for each line-divided image group G1 (t1) and the like generated by the integration unit 11 is image-integrated data obtained by imaging the same position in the inspection target T by the division output unit 12. Pixels that are always bright pixels whose luminance at the bright part l, the dark part d, and the boundary part b is equal to or greater than an arbitrary bright threshold, and pixels of the image integrated data obtained by imaging the same position on the inspection target T. The dark portion d and the boundary portion b are always dark pixels whose luminance is equal to or less than an arbitrary dark threshold, and the pixels of the image integration data obtained by imaging the same position on the inspection target T. The bright portion l, the dark portion d, and the boundary portion b Are divided and output for each pixel rich in light and dark changes whose width of change in luminance is equal to or greater than an arbitrary change threshold, so that the brightness of the image of the integrated image data for each line-divided image group G1 (t1) and the like. Part l, dark Since the appearance of at d and the boundary portion b is easily grasped, easily further recognize the type of the defect D of the inspection target T in the image integration data.

  In addition, according to the present embodiment, the integrated image data for each line-divided image group G1 (t1) and the like generated by the integration unit 11 is image-integrated data obtained by imaging the same position in the inspection target T by the division output unit 12. Pixels that are always bright pixels whose luminance at the bright part l, the dark part d, and the boundary part b is equal to or greater than an arbitrary bright threshold, and pixels of the image integrated data obtained by imaging the same position on the inspection target T. The dark portion d and the boundary portion b are always dark pixels whose luminance is equal to or less than an arbitrary dark threshold, and the pixels of the image integration data obtained by imaging the same position on the inspection target T. The bright portion l, the dark portion d, and the boundary portion b Pixels rich in light and dark changes in which the luminance change width is equal to or greater than an arbitrary change threshold are divided and output as colored portions R, G, and B colored in different colors. As a result, the appearance of the image of the integrated image data at the bright portion l, the dark portion d, and the boundary portion b can be easily grasped by the color for each line-divided image group G1 (t1) or the like. This makes it easier to recognize the type of the defect D.

  Further, according to the present embodiment, even if the image integrated data is obtained by integrating the pixel values of the image data, the line divided image is output by being divided for each line divided image group G1 (t1) and the like. Since the appearance of the image of the image integration data differs for each group G1 (t1) and the like, it becomes easy to recognize the type of the defect D of the inspection target T in the image integration data.

  Further, according to the present embodiment, even image integrated data in which pixel values of image data are integrated is divided and output for each predetermined threshold between the line divided image group G1 (t1) and the like. As a result, the appearance of the image of the image integration data differs for each predetermined threshold between the line-divided image group G1 (t1) and the like. Therefore, the type of the defect D of the inspection target T in the image integration data is changed. It becomes easy to recognize.

  Further, according to the present embodiment, the integration unit 11 captures the same position in the inspection target T for each of the line divided images DL1 (t1) and the like classified into the same line divided image group G1 (t1) and the like. Line division is performed by assigning differential values having positive and negative signs to the pixels according to the level of the luminance value of the pixel of the data, and integrating the differential values of the pixels of the image data obtained by imaging the same position in the inspection target T Since the integrated image data C1 (t1) and the like are generated for each image group G1 (t1) and the like, image data obtained by capturing the same position in the inspection target T for each line-divided image group G1 (t1) and the like by a simple calculation. The integrated image data C1 (t1) or the like in which the height of the pixel brightness with respect to the reference value is emphasized can be generated.

  Further, according to the present embodiment, the integration unit 11 captures the same position in the inspection target T for each of the line division images DL1 (t1) classified into the same line division image group G1 (t1) or the like. Are subjected to an emphasis process for emphasizing the luminance change of pixels adjacent to each other, and the values obtained by emphasizing the pixels of the image data obtained by imaging the same position in the inspection target T are integrated to obtain a line divided image group G1 (t1 ) To generate the image integration data C1 (t1) etc. for each, etc., the pixels adjacent to each other in the image data obtained by imaging the same position in the inspection target T for each line divided image group G1 (t1) etc. The integrated image data C1 (t1) and the like in which the luminance change is emphasized can be generated.

  Further, according to the present embodiment, the analysis unit 8 accumulates the result of machine learning related to the identification of the type of defect included in the image integration data for each of the line divided image groups G1 (t1) generated by the integration unit 11. The type of the defect D of the inspection target T is identified based on the processed data, but the image integration data in which the pixel values of the image data are integrated is the pixel of the image data for each line-divided image group G1 (t1) or the like. Are integrated, and therefore the defect type is identified based on the result of machine learning on the image integration data C1 (t1) or the like. Therefore, the identification accuracy of the type of the defect D of the inspection target T is identified. Can be improved.

  As mentioned above, although embodiment of this invention was described, this invention is implemented in various forms, without being limited to the said embodiment. For example, in the above embodiment, the case where the inspection target T is a film has been mainly described. However, the defect inspection system and the defect inspection method of the present invention are, for example, inspecting the filling amount of a liquid filled in a container in a production line. Can be applied to. By the defect inspection system 1 and the defect inspection method of the present embodiment, a defect such as whether the liquid does not reach a desired position in the container or whether the liquid does not exceed the desired position in the container is detected. be able to.

  Moreover, the defect inspection system 1 and the defect inspection method of this embodiment can be applied to appearance inspections such as cracks and scratches on glass products and the like in a production line. When the glass product has a defect such as a crack or a scratch, the defect can be extracted by utilizing the fact that the luminance is higher than that of other parts.

  DESCRIPTION OF SYMBOLS 1 ... Defect inspection system, 2 ... Light source, 3 ... Imaging part, 4 ... Conveyance part, 5 ... Image processing part, 6 ... Shading body, 7 ... Parallel light lens, 8 ... Analysis part, 9 ... Line division processing part, 10 DESCRIPTION OF SYMBOLS: Classification part, 11 ... Integration part, 12 ... Divided output part, 100 ... Convolutional neural network, 110 ... Input layer, 120 ... Hidden layer, 121, 123 ... Convolutional layer, 122 ... Pooling layer, 124 ... Fully connected layer, 130 ... output layer, T ... inspection object, X ... conveying direction, Y ... width direction, F (t1) ... two-dimensional image, l ... bright part, d ... dark part, b ... boundary part, D ... defect, DL1 (t1) ... line division image, L1 (t1) ... line, G1 (t1) ... line division image group, C1 (t1) ... image integration data, R, G, B ... coloring part, f ... weighting filter, n ... weighting coefficient, R: Reverse direction.

Claims (22)

A light source for irradiating the inspection object
An imaging unit that captures, every discrete time, a two-dimensional image of the light that is irradiated from the light source onto the inspection target and transmitted or reflected by the inspection target;
A transport unit that transports the inspection object relative to the light source and the imaging unit in a transport direction;
An image processing unit that processes image data of the two-dimensional image captured by the imaging unit;
With
The imaging unit
Capturing the two-dimensional image in which the luminance changes in a direction that matches the transport direction in the two-dimensional image;
The image processing unit
A line in which the two-dimensional image is divided into a plurality of lines arranged in parallel in the transport direction, and the lines at the same position in each of the two-dimensional images captured at the discrete time by the imaging unit are arranged in chronological order. A line division processing unit for processing the image data of the divided image;
A classification unit that classifies each of the line division images processed by the line division processing unit into two or more line division image groups according to a predetermined rule;
For each of the line-divided images classified into the same line-divided image group by the classification unit, the pixel values of the image data obtained by imaging the same position in the inspection object are integrated, and for each line-divided image group An integration unit for generating image integration data;
A division output unit that divides and outputs the image integration data for each line division image group generated by the integration unit according to a predetermined rule;
Having a defect inspection system. The two-dimensional image captured at discrete time by the imaging unit has a bright part, a dark part, and a boundary part between the bright part and the dark part,
The transport unit is
The defect inspection system according to claim 1, wherein the inspection object is conveyed relative to the imaging unit in the conveyance direction intersecting the bright part, the dark part, and the boundary part. The classification unit includes:
The line-divided image group of the line-divided image in which each of the line-divided images processed by the line-dividing processing unit is arranged in time-sequential order with the lines of the bright portion in each of the two-dimensional images; The line-divided image group of the line-divided image in which the lines in the dark part in each of the two-dimensional images are arranged in chronological order, and the line in which the line at the boundary in each of the two-dimensional images is arranged in chronological order The defect inspection system according to claim 2, wherein the defect inspection system classifies the divided images into the line divided image group. The two-dimensional image captured at discrete time by the imaging unit is located between the light source and the inspection target and shields a part of the light emitted from the light source to the inspection target. A light-shielding body that forms the bright part, the dark part, and the boundary part;
The transport unit is
4. The defect inspection system according to claim 2, wherein the inspection object is relatively transported in the transport direction intersecting the bright part, the dark part, and the boundary part with respect to the light source, the light shield, and the imaging unit. . The divided output unit includes:
The image integration data for each group of line-divided images generated by the integration unit is the pixels of the image integration data obtained by imaging the same position in the inspection target, and is the bright part, the dark part, and the boundary part. The pixels of the image integrated data obtained by imaging the same position in the inspection object, and the luminances at the bright part, the dark part, and the boundary part are arbitrary dark thresholds. The pixel of the image integration data obtained by imaging the same position in the inspection target and the pixel, and the width of the luminance change in the bright part, the dark part, and the boundary part is greater than or equal to an arbitrary change threshold The defect inspection system of any one of Claims 2-4 which divides | segments and outputs to each of the said pixel. The divided output unit includes:
The image integration data for each group of line-divided images generated by the integration unit is the pixels of the image integration data obtained by imaging the same position in the inspection target, and is the bright part, the dark part, and the boundary part. The pixels of the image integrated data obtained by imaging the same position in the inspection object, and the luminances at the bright part, the dark part, and the boundary part are arbitrary dark thresholds. The pixel of the image integration data obtained by imaging the same position in the inspection target and the pixel, and the width of the luminance change in the bright part, the dark part, and the boundary part is greater than or equal to an arbitrary change threshold The defect inspection system according to claim 5, wherein each of the pixels is divided and output as colored portions colored in different colors. The divided output unit includes:
The defect inspection system according to claim 1, wherein the image integration data for each line division image group generated by the integration unit is divided and output for each line division image group. The divided output unit includes:
The image integration data for each line division image group generated by the integration unit is divided and output for each predetermined threshold between the line division image groups. Defect inspection system described in. The integration unit
For each of the line-divided images classified into the same line-divided image group, a positive or negative sign is given to the pixel according to the level of the luminance value of the pixel of the image data obtained by imaging the same position in the inspection object. The integrated difference values of the pixels of the image data obtained by capturing the same position in the inspection target are integrated to generate image integrated data for each line-divided image group. The defect inspection system according to any one of 8. The integration unit
For each of the line-divided images classified into the same line-divided image group, an emphasis process is performed to emphasize the luminance change of the pixels adjacent to each other in the image data obtained by imaging the same position in the inspection object, and the inspection object The image integration data is generated for each of the line-divided image groups by integrating the values subjected to the enhancement processing of the pixels of the image data obtained by imaging the same position in the image. Defect inspection system described in.   An analysis unit for identifying the type of defect to be inspected based on data accumulated as a result of machine learning related to identification of the type of defect included in the integrated image data for each line-divided image group generated by the integration unit The defect inspection system according to any one of claims 1 to 10, further comprising: An irradiation process for irradiating the inspection object with light from the light source of the defect inspection system;
An imaging step of capturing, every discrete time, a two-dimensional image of the light that is irradiated on the inspection target from the light source and transmitted or reflected from the inspection target by the imaging unit of the defect inspection system;
A transporting step of transporting the inspection object relative to the light source and the imaging unit in a transporting direction by the transporting unit of the defect inspection system;
An image processing step of processing image data of the two-dimensional image captured in the imaging step by an image processing unit of the defect inspection system;
With
In the imaging step,
Capturing the two-dimensional image in which the luminance changes in a direction that matches the transport direction in the two-dimensional image;
In the image processing step,
A line in which the two-dimensional image is divided into a plurality of lines arranged in parallel in the transport direction, and the lines at the same position in each of the two-dimensional images captured at the discrete time by the imaging unit are arranged in chronological order. A line division processing step for processing the image data of the divided image;
A classification step of classifying each of the line division images processed in the line division step into two or more line division image groups according to a predetermined rule;
For each of the line-divided images classified into the same line-divided image group in the classification step, the pixel values of the image data obtained by imaging the same position in the inspection object are integrated, and for each line-divided image group An integration process for generating image integration data;
A division output step of dividing and outputting the image integration data for each of the line division image groups generated in the integration step according to a predetermined rule;
A defect inspection method. In the irradiation step,
The two-dimensional image captured at discrete time in the imaging step has a bright part, a dark part, and a boundary part between the bright part and the dark part,
In the conveying step,
The defect inspection method according to claim 12, wherein the inspection object is conveyed relative to the imaging unit in the conveyance direction intersecting the bright part, the dark part, and the boundary part. In the classification step,
The line-divided image group of the line-divided image in which the line-divided images processed in the line-dividing processing step are arranged in a time-series order in the lines of the bright portions in each of the two-dimensional images; The line-divided image group of the line-divided image in which the lines in the dark part in each of the two-dimensional images are arranged in chronological order, and the line in which the line at the boundary in each of the two-dimensional images is arranged in chronological order The defect inspection method according to claim 13, wherein the defect is classified into the line-divided image group of the divided images. In the irradiation step,
The bright image in the two-dimensional image captured at discrete times in the imaging step by a light-shielding body that is located between the light source and the inspection target and shields part of the light emitted from the light source to the inspection target. Part, the dark part and the boundary part,
In the conveying step,
The defect inspection method according to claim 13 or 14, wherein the inspection object is relatively transported in the transport direction intersecting the bright part, the dark part, and the boundary part with respect to the light source, the light shield, and the imaging unit. . In the divided output step,
The image integration data for each line-divided image group generated in the integration step is the pixel of the image integration data obtained by imaging the same position in the inspection target, and the bright portion, the dark portion, and the boundary portion. The pixels of the image integrated data obtained by imaging the same position in the inspection object, and the luminances at the bright part, the dark part, and the boundary part are arbitrary dark thresholds. The pixel of the image integration data obtained by imaging the same position in the inspection target and the pixel, and the width of the luminance change in the bright part, the dark part, and the boundary part is greater than or equal to an arbitrary change threshold The defect inspection method according to any one of claims 13 to 15, wherein the pixel is divided and output for each of the pixels. In the divided output step,
The image integration data for each line-divided image group generated in the integration step is the pixel of the image integration data obtained by imaging the same position in the inspection target, and the bright portion, the dark portion, and the boundary portion. The pixels of the image integrated data obtained by imaging the same position in the inspection object, and the luminances at the bright part, the dark part, and the boundary part are arbitrary dark thresholds. The pixel of the image integration data obtained by imaging the same position in the inspection target and the pixel, and the width of the luminance change in the bright part, the dark part, and the boundary part is greater than or equal to an arbitrary change threshold The defect inspection method according to claim 16, wherein each of the pixels is divided and output as colored portions colored in different colors. In the divided output step,
The defect inspection method according to claim 12, wherein the image integration data for each of the line division image groups generated in the integration step is divided for each line division image group and output. The divided output step includes
18. The image integration data for each of the line-divided image groups generated by the integration step is divided and output for each predetermined threshold value between the line-divided image groups. 18. The defect inspection method described in 1. In the integration process,
For each of the line-divided images classified into the same line-divided image group, a positive or negative sign is given to the pixel according to the level of the luminance value of the pixel of the image data obtained by imaging the same position in the inspection object. The difference value which has is provided, The said difference value of the said pixel of the said image data which imaged the same position in the said test object is integrated, The image integrated data are produced | generated for every said line division | segmentation image group. 20. The defect inspection method according to any one of 19 above. In the integration process,
For each of the line-divided images classified into the same line-divided image group, an emphasis process is performed to emphasize the luminance change of the pixels adjacent to each other in the image data obtained by imaging the same position in the inspection object, and the inspection object The image integration data is generated for each of the line-divided image groups by integrating the values subjected to the enhancement processing of the pixels of the image data obtained by imaging the same position in the image. The defect inspection method described in 1.   An analysis step for identifying the type of defect to be inspected based on data obtained by accumulating a result of machine learning related to identification of a type of defect included in the image integration data for each line-divided image group generated in the integration step The defect inspection method according to any one of claims 12 to 21, further comprising: