JP2019061317A - Image processing method, image processing device and inspection method - Google Patents

Abstract

To highly accurately specify an object area in which an image of an object exists in an original image, while capturing the original image of the object by an imaging unit.SOLUTION: A method include: an image acquisition step which acquires an original image of an object as an illumination switching image for each switching of illumination light, while switching the illumination light toward the object in multistage; an image processing step which performs edge emphasis processing which emphasizes an edge of the image of the object and acquires an edge emphasis image, for each illumination switching image acquired by the image acquisition step; an image generation step which generates a synthetic image obtained by adding the plurality of edge emphasis images; and an area specification step which specifies an object area from the synthetic image.SELECTED DRAWING: Figure 4

Description

  The present invention is an image processing technique for specifying a target area in which an image of an object is present in an original image obtained by imaging an object irradiated with illumination light from all or part of a plurality of illumination directions different from one another The present invention relates to an inspection method for inspecting an object using the image processing technology.

  An image of a workpiece included in an original image obtained by imaging the workpiece (hereinafter referred to as “inspection image” in order to inspect whether there is a defect such as a scratch on an inspection object such as an industrial part (hereinafter referred to as “work”) "" Is compared with a reference image obtained by imaging a workpiece not including defects in advance with an imaging unit such as a camera, and an inspection method for inspecting various ing. When this inspection method is adopted, it is important to extract an inspection image properly from an original image in order to improve inspection accuracy. That is, since the original image includes the background image in addition to the inspection image, an area (corresponding to the “target area” of the present invention) in the original image in which the inspection image exists is specified. It is necessary to exclude the background image from the original image.

  Thus, image processing which removes a background image and extracts an inspection image is proposed conventionally. For example, in the apparatus described in Patent Document 1, a background image not including a workpiece is registered in advance under various imaging conditions, and an original image obtained by imaging the workpiece is compared with a background image corresponding to the imaging condition. Then the background image is excluded and the inspection image is obtained.

Japanese Patent Laid-Open No. 2002-33957

  However, in the above-described conventional apparatus, a background image is acquired as a background material to acquire a background image whenever the photographing conditions change, such as the illumination intensity of the illumination and the aperture of the camera. That is, the background is fixed and the background image is static. On the other hand, when an industrial part or the like is used as a target of inspection, the work is transported to the inspection position by a table, a hand or the like, and imaged at the inspection position. As described above, in an environment in which various peripheral mechanical elements operate around the work, the background image is not constant but changes dynamically. Therefore, instead of preparing a background image in advance, an area (hereinafter referred to as “target area”) including an image of a work is specified from an original image obtained by imaging the work to be inspected, and the specified target area It is desirable to extract examination images based on

  The present invention has been made in view of the above problems, and an image processing technique capable of specifying with high accuracy a target area in which an image of a target is present in the original image while capturing the target image by the imaging unit. And an inspection method capable of inspecting an object with high accuracy using the image processing technology.

  The first aspect of the present invention is an image for specifying a target area in which an image of a target is present in an original image obtained by imaging the target illuminated with illumination light from all or part of a plurality of different illumination directions. A processing method, comprising: an image acquiring step of acquiring an original image of an object as an illumination switching image at each switching of the illumination light while switching illumination light to the object in multiple stages; and illumination switching acquired in the image acquiring step An image processing step of acquiring an edge-emphasized image by performing edge emphasis processing for emphasizing an edge of an image of an object for each image, an image generation step of generating a composite image obtained by adding a plurality of edge-emphasized images, and combining And a region specifying step of specifying a target region from an image.

  Further, according to the second aspect of the present invention, a target area in which an image of an object is present is specified in an original image obtained by imaging the object irradiated with illumination light from all or a part of a plurality of different illumination directions. An image acquisition unit configured to acquire an original image of an object as an illumination switching image each time the illumination light is switched while switching the illumination light to the object in multiple stages, and an object for each illumination switching image An image processing unit for obtaining an edge-emphasized image by performing edge enhancement processing for emphasizing an edge of an image of an object, an image generation unit for generating a composite image obtained by adding a plurality of edge-emphasized images, and specifying a target region from the composite image And an area specifying unit.

  Furthermore, a third aspect of the present invention is an inspection method for inspecting an object, comprising the steps of: specifying a target area in which an image of the target is present by the image processing method; And an inspection step of inspecting the object based on the image of the object extracted by the extraction step.

  In the invention configured as described above, the illumination switching image is acquired while switching the illumination light, and the edge enhancement processing is performed on each illumination switching image to acquire the edge-emphasized image. As described above, the edge-emphasized image is obtained while irradiating the illumination light in various illumination modes, and in the composite image generated by adding them, the edge of the image of the object becomes clear. Then, the target area is accurately identified from the composite image having such characteristics.

  As described above, according to the present invention, a plurality of illumination switching images are acquired while switching the illumination light to the object in multiple stages, and an edge-emphasized image obtained by performing edge enhancement processing on each illumination switching image is added. The target area is specified based on the composite image generated together. Therefore, the target area can be identified with high accuracy. Further, the image of the object can be extracted using the target region thus identified, and the object can be inspected with high accuracy by inspecting the object based on the image.

It is a figure showing composition of an inspection device equipped with one embodiment of an image processing device concerning the present invention. It is a top view which shows the main body of the inspection apparatus shown in FIG. It is a flow chart which shows a flow of inspection processing in an inspection device. It is a flowchart which shows the identification process of the object area | region performed in a test process. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by the upper imaging part, and an edge emphasis image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by the upper imaging part, and an edge emphasis image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by the upper imaging part, and an edge emphasis image. It is a figure which shows the test | inspection image extracted from the synthetic | combination image produced from the edge emphasis image shown in FIGS. 5-7, and the said synthetic | combination image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by an oblique imaging part and an edge emphasis image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by an oblique imaging part and an edge emphasis image. It is a figure which shows the test | inspection image extracted from the synthetic | combination image produced from the edge emphasis image shown to FIG. 9 and FIG. 10, and the said synthetic | combination image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by a side imaging part and an edge emphasis image. It is a figure which shows an example of the illumination switching image at the time of imaging a workpiece | work by a side imaging part and an edge emphasis image. It is a figure which shows the synthetic | combination image produced from the edge emphasis image shown to FIG. 12 and FIG. 13, and the test | inspection image extracted from the said synthetic | combination image.

  FIG. 1 is a view showing the configuration of an inspection apparatus equipped with an image processing apparatus according to an embodiment of the present invention. FIG. 2 is a plan view showing the main body of the inspection apparatus shown in FIG. The inspection apparatus 1 is an apparatus that inspects the workpiece W by comparing an inspection image obtained by imaging, for example, a metal workpiece W manufactured by forging or casting with a reference image acquired in advance.

  As shown in FIG. 1, the inspection apparatus 1 includes a main body 11 and a control unit 12 configured by a computer. The main body 11 includes a stage 2, an imaging unit 3, and a light source unit 4. The work W is placed on the stage 2. The main body 11 is provided with a light shielding cover (not shown) for preventing external light from reaching the stage 2, and the stage 2, the imaging unit 3 and the light source unit 4 are provided in the light shielding cover.

  As shown in FIGS. 1 and 2, the imaging unit 3 includes one upper imaging unit 31, eight oblique imaging units 32, and eight side imaging units 33. In FIG. 2, the upper imaging unit 31 is not shown (the same applies to the upper light source unit 41 described later). The upper imaging unit 31 is disposed on a central axis J1 extending vertically upward from the center of the stage 2 above the stage 2 as shown in FIG. The original image captured from above is output to the control unit 12.

  As shown in FIG. 2, when viewing the main body 11 from the upper side to the lower side (that is, when the main body 11 is viewed in plan), the eight oblique imaging units 32 are arranged around the stage 2 There is. The eight oblique imaging units 32 are arranged at an angular interval (pitch) of 45 ° in the circumferential direction about the central axis J1. In a plane including the imaging optical axis K2 of each oblique imaging unit 32 and the central axis J1 (see FIG. 1), an angle θ2 formed by the imaging optical axis K2 and the central axis J1 is approximately 45 °. An original image obtained by imaging the work W on the stage 2 obliquely from the top can be acquired by each oblique imaging unit 32.

  Similar to the eight oblique imaging units 32, the eight side imaging units 33 are also arranged around the stage 2 when the main body 11 is viewed in plan. The eight side imaging units 33 are arranged at an angular interval of 45 ° in the circumferential direction about the central axis J1. In a plane including the imaging optical axis K3 of each side imaging unit 33 and the central axis J1, an angle θ3 formed by the imaging optical axis K3 and the central axis J1 is approximately 90 °. An original image obtained by imaging the work W on the stage 2 from the side can be acquired by each side imaging unit 33. The upper imaging unit 31, the oblique imaging unit 32, and the side imaging unit 33 have, for example, a charge coupled device (CCD), a complementary metal oxide semiconductor (CMOS), and the like, and a multi-gradation image is obtained. The upper imaging unit 31, the oblique imaging unit 32, and the side imaging unit 33 are supported by a support (not shown).

  The light source unit 4 includes one upper light source unit 41, eight oblique light source units 42, and eight side light source units 43. The upper light source unit 41 is a light source unit in which a plurality of LEDs (light emitting diodes) are arranged in a ring shape centering on the central axis J1. The ring-shaped upper light source unit 41 is fixed to the upper imaging unit 31 so as to surround the periphery of the upper imaging unit 31. The illumination light can be irradiated by the upper light source unit 41 along the direction parallel to the central axis J1 from directly above the work W on the stage 2.

  When the main body 11 is viewed in plan, eight oblique light source units 42 are disposed around the stage 2. The eight oblique light source parts 42 are arranged at an angular interval of 45 ° in the circumferential direction centering on the central axis J1. Each of the oblique light source units 42 is a light source unit in which a plurality of LEDs are arranged in a bar shape extending in a tangential direction of a circumference centered on the central axis J1. The line connecting the center of the exit surface of each oblique light source unit 42 and (the center of) the work W is referred to as the “diagonal illumination axis”. The oblique illumination axis of the oblique light source unit 42 and the central axis J1 In the included plane, the angle between the oblique illumination axis and the central axis J1 is about 45 °. In each of the oblique light source units 42, illumination light can be irradiated to the work W on the stage 2 from obliquely above along the oblique illumination axis. In the inspection apparatus 1, the oblique light source units 42 are respectively fixed to the oblique imaging unit 32.

  When the main body 11 is viewed in a plan view, eight side light source units 43 are disposed around the stage 2. The eight side light source units 43 are arranged at an angular interval of 45 ° in the circumferential direction about the central axis J1. Each side light source unit 43 is a light source unit in which a plurality of LEDs are arranged in a bar shape extending in a tangential direction of a circumference centered on the central axis J1. Similarly to the oblique light source unit 42, when a line connecting the center of the emission surface of each side light source unit 43 and the work W is referred to as “side illumination axis”, the side illumination axis of the side light source unit 43 and In the plane including the central axis J1, the angle between the side illumination axis and the central axis J1 is approximately 90 degrees. In each side light source unit 43, illumination light can be irradiated to the work W on the stage 2 from the side along the side illumination axis. In the inspection apparatus 1, the side light source units 43 are fixed to the side imaging units 33 respectively.

  As described above, in the present embodiment, 17 light source units (= 1 upper light source unit 41 + 8 oblique light source units 42 + 8 side light source units 43) are arranged so as to surround the periphery of the stage 2 in a dome shape. The workpiece W can be illuminated with illumination light from all or a part of 17 illumination directions different from one another. For example, the distance between the upper imaging unit 31 and the upper light source unit 41 and the work W is about 55 cm (centimeter). The distance between the oblique imaging unit 32 and the oblique light source unit 42 and the work W is about 50 cm, and the distance between the side imaging unit 33 and the side light source unit 43 and the work W is about 40 cm. is there. In the upper light source unit 41, the oblique light source unit 42, and the side light source unit 43, light sources of types other than LEDs may be used.

  In the present embodiment, a control unit 12 is provided to control each part of the main body 11 and to inspect the workpiece W by comparing the inspection image of the workpiece W with the reference image as described later. The control unit 12 includes an arithmetic processing unit 5 configured by a CPU (Central Processing Unit), reference image data 61, illumination switching image data 62, edge-emphasized image data 63, composite image data 64, and a target area, which will be described in detail later. The storage unit 6 stores various data such as the data 65 and the inspection image data 66, a program, and the like. The arithmetic processing unit 5 controls each part of the apparatus according to the above program to obtain a plurality of illumination switching images while switching the illumination light to the workpiece W in multiple stages, and performs smoothing processing on each illumination switching image and Edge emphasis processing is performed to obtain a plurality of edge-emphasized images, the edge-emphasized images are added together to generate a composite image, a target area is specified based on the composite image, and an image of the work W, ie, based on the target area The inspection image is extracted, and the inspection image and the reference image are compared to inspect the workpiece W. As described above, the arithmetic processing unit 5 functions as the image acquisition unit 51, the image processing unit 52, the image generation unit 53, the target area identification unit 54, the image extraction unit 55, and the inspection unit 56. Hereinafter, identification of a target area from an original image including an image of the work W and inspection of the work W will be described with reference to FIGS. 3 to 14.

  FIG. 3 is a flowchart showing the flow of inspection processing in the inspection apparatus. Further, FIG. 4 is a flowchart showing a process of specifying a target area performed in the inspection process. First, the workpiece W to be inspected is placed on the stage 2 (step S1). A holding unit (not shown) that conforms to the shape, size, etc. of the work W is provided on the stage 2, and is positioned at a predetermined position on the stage 2 with the main surface W 1 of the work W directed vertically upward. Hold the work W. Subsequently, the control unit 12 acquires imaging setting information on the workpiece W on the stage 2 based on an input by the operator or the like (step S2). Here, the imaging setting information includes an imaging unit (hereinafter referred to as a “selected imaging unit”) used in the imaging unit 3 and a light source unit which lights in the light source unit 4 when the selected imaging unit acquires a photographed image. Show. For example, in the case where the workpiece W is imaged from three imaging directions in the upper, oblique, and side directions to acquire three original images and the workpiece W is inspected (hereinafter, referred to as “specific example”), the imaging setting information is The following information may be included.

The upper imaging unit 31 is used as a selective imaging unit, and all the oblique light source units 42 are turned on as a light source unit in order to illuminate the work W from the illumination direction suitable for it.
An oblique light source unit for illuminating the work W from a suitable illumination direction while using one oblique imaging unit 32a (see FIG. 2) of the eight oblique imaging units 32 as a selective imaging unit Light part of 42 as a light source,
A side light source unit is used to illuminate the work W from a suitable illumination direction while using one side imaging unit 33a (see FIG. 2) of the eight side imaging units 33 as a selected imaging unit Light part of 43 as a light source,
It is needless to say that the type and number of imaging units to be used and the lighting mode of the light source unit are not limited to this, and may be appropriately changed according to the type of the work W and the like.

  When the imaging setting information for the workpiece W is acquired, the arithmetic processing unit 5 executes a process of specifying a target area prior to imaging and inspection of the workpiece W based on the imaging setting information (step S3). First, the arithmetic processing unit 5 selects one of the selected imaging units (corresponding to the upper imaging unit 31, the oblique imaging unit 32a, and the side imaging unit 33a in the above specific example) that has not completed the specification of the target area. A set is determined (step S31). In addition, the arithmetic processing unit 5 sets a light source unit (hereinafter referred to as a "lighting light source unit") and a plurality of lighting patterns to be turned on when imaging the work W with the selected imaging unit determined in step S31 (step S32). Although the lighting pattern will be described later using a specific example, the work W can be illuminated with a plurality of illumination modes by switching these lighting patterns.

  The arithmetic processing unit 5 switches to one lighting pattern among the plurality of lighting patterns, and in a state where the desired lighting light source unit is selectively lit, the original image of the work W captured by the selected imaging unit is switched to the illumination switching image Image data (illumination switching image data 62) is stored in the storage unit 6 (step S33). Subsequently, the arithmetic processing unit 5 performs smoothing processing on the illumination switching image data 62 to remove noise components included in the image, and then performs edge emphasizing processing to emphasize the edge of the image of the workpiece W. The edge-emphasized image is acquired, and the image data (edge-emphasized image data 63) is stored in the storage unit 6 (step S34). The image acquisition process (step S33) and the image processing process (step S34) are repeated until it is determined by the arithmetic processing unit 5 that the entire lighting pattern has been completed ("YES" in step S35). By this repeated operation, while the illumination light is irradiated to the work W in a plurality of illumination modes, an edge-emphasized image for each illumination mode is obtained.

  In the next step S36, the arithmetic processing unit 5 adds together the plurality of edge-emphasized images obtained in this way to generate a composite image, and stores composite image data 64 in the storage unit 6 (image generation step). Furthermore, the arithmetic processing unit 5 subjects the composite image to a binarization process, and the target area (in which the image of the work W exists in the original image of the work W captured by the selected imaging unit determined in step S31) The code | symbol OR in FIG.8, FIG.11 and FIG.14 demonstrated in (step S37: area | region identification process) is specified. Then, the arithmetic processing unit 5 determines whether or not the target region has been specified for all of the selected imaging units (step S38), and while determining that all have not been completed, the process returns to step S31 and the above series of processes ( The image acquisition step + the image processing step + the image generation step + the area specifying step is repeated to specify the target area for each selected imaging unit, and the target area data 65 is stored in the storage unit 6.

  Here, in order to further deepen the understanding of the process of specifying the target area, the process of specifying the target area will be specifically described based on the specific example. In the above specific example, among the 17 imaging units, three of the upper imaging unit 31, the oblique imaging unit 32a, and the side imaging unit 33a are selected as the selection imaging unit.

  When the upper imaging unit 31 is determined to be "one selected imaging unit" in step S31, as shown in FIGS. 5 to 7, all eight oblique light source units 42 are selected as lighting light source units. A total of eight lighting patterns (A-1) to (A-8) for setting the respective oblique light source units 42 one by one are set (step S32). In these drawings (and in FIG. 9, FIG. 10, FIG. 12 and FIG. 13 described later), dots are attached to clearly indicate the light source unit 42 to be lit. Then, when switched to the lighting pattern (A-1), the lighting switching image SG (A-1) is acquired in the lighting mode in which only the lighting light source unit 42a is lighted (step S33). Then, the edge enhanced image EG (A-1) is acquired by the smoothing process and the edge enhancement process on the illumination switching image SG (A-1) (step S34). Further, each time the lighting patterns (A-2) to (A-8) are sequentially switched, the edge enhanced images EG (A-2) to EG (A-8) are acquired in the same manner as described above. The eight edge-weighted images EG (A-1) to EG (A-8) thus acquired are added together to generate a composite image as shown in the upper part of FIG. 8 (step S36). By adding binarization processing to the composite image, as shown in the lower part of FIG. 8, the target region OR surrounded by the boundary line BL is determined (step S37). The image shown in the lower part of FIG. 8 shows an inspection image obtained by the extraction processing described below.

  When the oblique imaging unit 32a (see FIG. 2) is determined to be “one selected imaging unit” in step S31, as shown in FIGS. 9 and 10, the eight oblique light source units 42 are selected. Four lighting patterns (B-1) to (B) are selected in which five of the oblique light source parts 42b to 42f and 42h are selected as the lighting light source parts and one each of the oblique light source parts 42c to 42e and 42h are lit. A total of five lighting patterns of -4) and a lighting pattern (B-5) for simultaneously lighting the five oblique light source parts 42b to 42f are set (step S32). Then, when the lighting pattern (B-1) is switched, the lighting switching image SG (B-1) is acquired in the lighting mode in which only the lighting light source unit 42h lights (step S33). Then, the edge enhanced image EG (B-1) is obtained by the smoothing process and the edge enhancement process on the illumination switching image SG (B-1) (step S34). Further, each time the lighting patterns (B-2) to (B-5) are sequentially switched, the edge enhanced images EG (B-2) to EG (B-5) are acquired in the same manner as described above. By combining the six edge-emphasized images EG (B-1) to EG (B-5) thus acquired, a composite image as shown in the upper part of FIG. 11 is generated (step S36). By adding binarization processing to the composite image, as shown in the lower part of FIG. 11, the target area OR surrounded by the boundary line BL is determined (step S37). The image shown in the lower part of FIG. 11 shows an inspection image obtained by the extraction processing described below.

  Furthermore, when the oblique imaging unit 33a (see FIG. 2) is determined to be “one selected imaging unit” in step S31, as shown in FIGS. 12 and 13, the eight side light source units 43 are selected. Three lighting patterns (C-1) to (C-3) in which five of the side light source units 43b to 43f are selected as lighting light source units and one side light source unit 43c to 43e are turned on A total of four lighting patterns are set with the lighting pattern (C-4) for simultaneously lighting the five side light source parts 43b to 43f (step S32). Then, when switched to the lighting pattern (C-1), the illumination switching image SG (C-1) is acquired in the illumination mode in which only the side light source unit 43d is lit (step S33). Then, the edge enhanced image EG (C-1) is obtained by the smoothing process and the edge enhancement process on the illumination switching image SG (C-1) (step S34). Further, each time the lighting patterns (C-2) to (C-4) are sequentially switched, the edge-emphasized images EG (C-2) to EG (C-4) are acquired in the same manner as described above. By combining the four edge-emphasized images EG (C-1) to EG (C-4) acquired in this manner, a composite image as shown in the upper part of FIG. 14 is generated (step S36). By adding binarization processing to the composite image, as shown in the lower part of FIG. 14, the target region OR surrounded by the boundary line BL is determined (step S37). The image shown in the lower part of FIG. 14 shows an inspection image obtained by the extraction processing described below.

  Next, returning to FIG. 3, the description will be continued. When identification of the target area OR (see FIG. 8, FIG. 11, and FIG. 14) is completed, the arithmetic processing unit 5 reads the composite image data 64 and the target region data 65 from the storage unit 6, Extract (step S4: extraction step). That is, an image other than the target area OR in the composite image is excluded as the background image by the area threshold and the image of the work W, that is, the inspection image is extracted (see the lower part of FIG. 8, FIG. 11 and FIG. 14) 66 is stored in the storage unit 6. Then, the arithmetic processing unit 5 reads from the storage unit 6 the data of the reference image of the work W stored in advance in the storage unit 6, that is, the reference image data 61 (step S5: reading step). The reference image is a photographed image obtained by imaging the workpiece W having no defect such as a scratch based on the imaging setting information, and is acquired before the inspection and stored in the storage unit 6. Finally, the arithmetic processing unit 5 compares the inspection image with the reference image to inspect whether there is a defect such as a flaw in the work W (step S6: inspection step).

  As described above, according to the present embodiment, the illumination switching image is acquired while switching the illumination light, and the edge emphasis processing is performed on each illumination switching image to acquire the edge-emphasized image, and further, their edge emphasis is performed. Images are added together to generate a composite image. Therefore, the edge of the work W becomes clear in the composite image, and then the target area where the image of the work W (inspection image) is present can be identified with high accuracy. In addition, since the inspection image is extracted using the target area thus identified, the workpiece W can be inspected with high accuracy.

  In the above embodiment, the work W corresponds to an example of the "object" of the present invention. Further, the arithmetic processing unit 5 functions as the "image processing apparatus" of the present invention.

  The present invention is not limited to the above-described embodiment, and various modifications can be made other than the above without departing from the scope of the invention. For example, in the present embodiment, the extraction step, the reading step, and the inspection step are performed after the identification of the target region is completed for all selected imaging units, but the execution timing of the extraction step, the reading step, and the inspection step is limited thereto. For example, the identification process of the target area and the extraction process may be performed continuously for each selected imaging unit. In addition, identification of a target area, an extraction process, a reading process, and an inspection process may be performed for each selected imaging unit.

  Further, in the above embodiment, the present invention is applied to an inspection apparatus which selects three imaging units 31 and the oblique imaging unit 32a side imaging unit 33a among the 17 imaging units as a selection imaging unit. The type and number of imaging units are not limited to this. In addition, although the present invention is applied to an inspection apparatus provided with 17 imaging units and 17 light source units, the number of imaging units and light source units and the type of the apparatus are limited to the above embodiment. The present invention can be generally applied to an apparatus generally provided with an imaging unit for imaging an object irradiated with illumination light from all or part of a plurality of illumination directions different from one another.

  The present invention relates generally to an image processing technique for specifying a target area in which an image of an object is present in an original image obtained by imaging the object irradiated with illumination light from all or part of a plurality of different illumination directions. And it can apply to the inspection method which inspects an object using the image processing technique concerned.

1 ... inspection device 5 ... arithmetic processing unit (image processing device)
31 ... upper imaging unit 32 ... oblique imaging unit 33 ... side imaging unit 41 ... upper light source unit 42 ... oblique light source unit 43 ... side light source unit 51 ... image acquisition unit 52 ... image processing unit 53 ... image generation unit 54 ... Target area identification unit 55 ... Image extraction unit 56 ... Inspection unit OG ... Target region SG ... Illumination switching image W ... Work (object)

In the present embodiment, the control unit 12 is provided for inspecting the workpiece W by comparing the reference image inspection image of the workpiece W as described later and controls the respective units of the main body 11. The control unit 12 includes an arithmetic processing unit 5 configured by a CPU (Central Processing Unit), reference image data 61, illumination switching image data 62, edge-emphasized image data 63, composite image data 64, and a target area, which will be described in detail later. The storage unit 6 stores various data such as the data 65 and the inspection image data 66, a program, and the like. The arithmetic processing unit 5 controls each part of the apparatus according to the above program to obtain a plurality of illumination switching images while switching the illumination light to the workpiece W in multiple stages, and performs smoothing processing on each illumination switching image and Edge emphasis processing is performed to obtain a plurality of edge-emphasized images, the edge-emphasized images are added together to generate a composite image, a target area is specified based on the composite image, and an image of the work W, ie, based on the target area The inspection image is extracted, and the inspection image and the reference image are compared to inspect the workpiece W. As described above, the arithmetic processing unit 5 functions as the image acquisition unit 51, the image processing unit 52, the image generation unit 53, the target area identification unit 54, the image extraction unit 55, and the inspection unit 56. Hereinafter, identification of a target area from an original image including an image of the work W and inspection of the work W will be described with reference to FIGS. 3 to 14.

Next, returning to FIG. 3, the description will be continued. When the specification of the target area OR ( see FIG. 8, FIG. 11, and FIG. 14) is completed, the arithmetic processing unit 5 reads the composite image data 64 and the target region data 65 from the storage unit 6, and based on them, the inspection image is composite image (Step S4: extraction step). That is, an image other than the target area OR in the composite image is excluded as the background image by the area threshold and the image of the work W, that is, the inspection image is extracted (see the lower part of FIG. 8, FIG. 11 and FIG. 14) 66 is stored in the storage unit 6. Then, the arithmetic processing unit 5 reads from the storage unit 6 the data of the reference image of the work W stored in advance in the storage unit 6, that is, the reference image data 61 (step S5: reading step). The reference image is a photographed image obtained by imaging the workpiece W having no defect such as a scratch based on the imaging setting information, and is acquired before the inspection and stored in the storage unit 6. Finally, the arithmetic processing unit 5 compares the inspection image with the reference image to inspect whether there is a defect such as a flaw in the work W (step S6: inspection step).

Claims (6)

An image processing method for specifying a target area in which an image of an object is present in an original image obtained by imaging an object irradiated with illumination light from all or part of a plurality of different illumination directions.
An image acquisition step of acquiring an original image of the object as an illumination switching image each time the illumination light is switched while switching the illumination light to the object in multiple stages;
An image processing step of acquiring an edge-emphasized image by performing edge emphasis processing for emphasizing an edge of the image of the object for each of the illumination switching images acquired in the image acquisition step;
An image generation step of generating a composite image in which the plurality of edge-emphasized images are added together;
An area specifying step of specifying the target area from the composite image. The image processing method according to claim 1,
The image processing step is an image processing method of performing the edge enhancement processing after performing the smoothing processing on the illumination switching image to obtain the edge enhanced image. The image processing method according to claim 1 or 2, wherein
In the image processing method, the extraction step performs binarization processing on the composite image to specify the target area. The image processing method according to any one of claims 1 to 3, wherein
A plurality of imaging units are provided to image the object from different imaging directions,
The target area where the image of the target exists in the original image captured by the imaging unit by performing the image acquisition step, the image processing step, the image generation step, and the area specifying step for each of the imaging units Image processing method to identify. An image processing apparatus for specifying a target area in which an image of an object is present in an original image obtained by imaging the object irradiated with illumination light from all or part of a plurality of different illumination directions.
An image acquisition unit that acquires an original image of the object as an illumination switching image at each switching of the illumination light while switching the illumination light to the object in multiple stages;
An image processing unit for performing an edge enhancement process for enhancing an edge of the image of the object for each of the illumination switching images to obtain an edge enhanced image;
An image generation unit configured to generate a composite image obtained by adding together the plurality of edge-emphasized images;
And a region specifying unit for specifying the target region from the composite image. An inspection method for inspecting an object,
A process of specifying a target area in which an image of the target exists by the image processing method according to any one of claims 1 to 4.
An extraction step of excluding an image other than the target area among the composite image to extract an image of the target;
And D. an inspection step of inspecting the object based on the image of the object extracted in the extraction step.