JP2022146141A - Visual inspection method and visual inspection device - Google Patents

Abstract

To improve inspection accuracy when performing a visual inspection on an inspection object using a color image.SOLUTION: A visual inspection method comprises the steps of: acquiring an image of a workpiece W irradiated with light; acquiring a first image being a monochromatic light image in a first color and a second image being a monochromatic light image in a second color being a different color from the first color from the acquired image of the workpiece W; combining the image based on the first image with the image based on the second image by adding the respective gradation values of a plurality of pixels constituting each image to each other; extracting a portion whose gradation value is within a prescribed range from the obtained composite image; and determining the presence/absence of a defect C on the basis of the extracted portion.SELECTED DRAWING: Figure 6

Description

The technology disclosed herein belongs to the technical field related to visual inspection methods and visual inspection apparatuses.

2. Description of the Related Art Conventionally, there has been known a visual inspection method for inspecting an inspection object such as an electronic component for defects such as chipping and adhesion of foreign matter based on an image obtained by imaging the inspection object. Conventionally, inspections based on black-and-white images have been the mainstream, but in recent years, more detailed inspections using color images have been considered.

For example, in the appearance inspection apparatus described in Patent Document 1, each of a plurality of captured images is spectrally processed into a plurality of color components to generate a plurality of spectral images, and the generated spectral images are divided into distribution ratios according to the purpose of inspection. generates a composite image synthesized in , and determines the appearance of the inspection object based on the generated composite image.

JP 2019-124640 A

By the way, when the defect is colored, there is a difference in the contrast of the defect portion in each of the spectral images. When the contrast of the defect portion of each spectral image is low, if the distribution ratio is determined for each spectral image and synthesized, the contrast information may be degraded. Therefore, there is room for improvement from the viewpoint of improving the inspection accuracy of appearance inspection.

The technique disclosed herein has been made in view of the above points, and its purpose is to improve the inspection accuracy when performing the appearance inspection of an inspection object using a color image.

In order to solve the above problems, the technology disclosed herein is directed to a visual inspection method for inspecting defects formed on the surface of an inspection object, and includes an irradiation step of irradiating the inspection object with light; a first image that is a monochromatic light image of a first color from the acquired image of the inspection object; and a color different from the first color. a monochromatic light image acquiring step of acquiring a second image that is a monochromatic light image of a second color; an image synthesizing step of synthesizing each tone value by adding it up for each corresponding pixel; and a judging step of judging the presence or absence of the defect based on the synthetic image obtained in the image synthesizing step. The step extracts a portion having a gradation value within a predetermined range from the synthesized image obtained in the image synthesizing step, and determines the presence or absence of the defect based on the extracted portion. did.

According to this configuration, since the image is synthesized by adding the gradation values, the difference between the gradation values of the defect candidate and its periphery becomes large, and the contrast around the defect becomes large. In particular, by using monochromatic light images of different colors, the difference in gradation value is caused by a defect, unlike the case where the gradation value of one image is doubled to increase the difference in gradation value. It can be determined with high accuracy that there is In addition, by extracting a portion whose gradation value is within a predetermined range, the periphery of the defect can be emphasized. Thereby, the inspection accuracy of the appearance inspection can be improved.

In one embodiment of the appearance inspection method, the light irradiated in the irradiation step is white light, and the monochromatic light image obtaining step obtains an image of two colors of RGB from the image obtained in the imaging step. as the first image and the second image.

That is, if it is an image of an inspection object irradiated with white light, an image of each color of RGB can be generated from the image. This eliminates the need to obtain an image in which the inspection object is irradiated with each color of RGB, thereby shortening the inspection time.

In the above-described embodiment, the monochromatic light image acquiring step uses the monochromatic light image of a color having the highest contrast around the defect candidate among RGB as the first image, and the color having the smallest contrast around the defect candidate among RGB. using the monochromatic light image as the second image, further comprising a reversed image obtaining step of obtaining a reversed image by reversing the gradation value in the second image before the image synthesizing step, wherein the image synthesizing step The step may be a step of synthesizing the first image and the reverse image.

According to this configuration, by synthesizing an inverted image obtained by inverting the gradation value of the monochromatic light image of the color with the lowest contrast around the defect candidate with the first image, the gradation value of the defect portion is almost affected. gradation value of the background portion can be made as small as possible. Thereby, the contrast around the defect can be enhanced. As a result, it is possible to further improve the inspection accuracy of the appearance inspection.

In the one embodiment, before the image synthesizing step, the average value of the gradation values of the pixels of the first image and the average value of the gradation values of the pixels of the second image are adjusted to be the same. and further comprising a gradation value adjustment step of correcting the gradation value of each pixel of the first image and the gradation value of each pixel of the second image, wherein the image synthesizing step is performed so that the gradation value of each pixel is It may be a step of synthesizing the first image after each adjustment and the second image after each adjustment of the gradation value of each pixel.

According to this configuration, even if the light receiving sensitivity of the camera used in the imaging process is different for RGB, or if the wavelength balance of RGB of the lighting device is different, the average value of the gradation values of the first image and the second image These effects can be reduced by aligning the Thereby, the contrast around the defect can be easily emphasized, and the inspection accuracy of the appearance inspection can be further improved.

Another aspect of the technology disclosed herein is directed to a visual inspection apparatus for inspecting defects formed on a surface of an inspection object, and includes an irradiation unit for irradiating the inspection object with light, and an imaging unit that acquires an image of the inspection object; and an inspection unit that inspects whether or not there is a defect from the acquired image of the inspection object, wherein the inspection unit performs the inspection acquired by the imaging unit. obtaining a first image, which is a monochromatic light image of a first color, and a second image, which is a monochromatic light image of a second color different from the first color, from the image of the object; generating a composite image by combining the base image and the image based on the second image by adding up the tone values of the plurality of pixels constituting each image for each corresponding pixel, and from the composite image, A portion having a gradation value within a predetermined range is extracted, and the presence or absence of the defect is determined based on the extracted portion.

With this configuration as well, the difference in gradation value around the defect can be increased, and the contrast around the defect can be enhanced. Thereby, the inspection accuracy of the appearance inspection can be improved.

In the appearance inspection apparatus, the light emitted by the irradiation unit is white light, and the inspection unit converts images of two colors of RGB from the image acquired by the imaging unit into the first image and the A configuration in which the image is obtained as the second image may also be used.

According to this configuration, an image of each color of RGB can be generated from the acquired image. This eliminates the need to obtain an image in which the inspection object is irradiated with each color of RGB, thereby shortening the inspection time.

Another aspect of the visual inspection method in the technique disclosed herein is directed to a visual inspection method for inspecting defects formed on the surface of an inspection object, and includes an irradiation step of irradiating the inspection object with light; an imaging step of acquiring one or more images of the inspected object irradiated with, and when the image is one, for a plurality of monochromatic light images acquired from the one image, each Each monochromatic light image is synthesized by summing up the respective gradation values of a plurality of pixels constituting the monochromatic light image for each corresponding pixel, and when there are a plurality of said images, the plurality of monochromatic light images constituting the respective images are synthesized. an image synthesizing step of synthesizing each image by adding up the respective gradation values of the pixels for each corresponding pixel; and a determining step of determining the presence or absence of the defect based on the synthesized image obtained in the image synthesizing step; wherein the determining step is a step of extracting a portion having a gradation value within a predetermined range from the synthesized image obtained in the image synthesizing step, and determining the presence or absence of the defect based on the extracted portion. It was configured as follows.

With this configuration as well, the difference in gradation value around the defect can be increased, and the contrast around the defect can be enhanced. Thereby, the inspection accuracy of the appearance inspection can be improved.

As described above, according to the technology disclosed herein, it is possible to improve the inspection accuracy when performing a visual inspection of an inspection object using a color image.

FIG. 1 is a schematic diagram of a visual inspection apparatus according to Embodiment 1. FIG. FIG. 2 is a schematic diagram showing the periphery of the camera of the visual inspection apparatus. FIG. 3 is a block diagram showing the control system of the visual inspection apparatus. FIG. 4 is a schematic diagram showing a state in which a monochromatic light image is generated from an image captured by a camera. FIG. 5 is a diagram schematically showing the image synthesizing process. FIG. 6 is a diagram showing how an image for determination is acquired from a synthesized image. FIG. 7 is a flow chart showing the processing operation of appearance inspection by the controller. FIG. 8 is a diagram schematically showing a reverse image generation process in the visual inspection apparatus according to the second embodiment. FIG. 9 is a diagram showing how an image for judgment is acquired from a synthesized image in the appearance inspection apparatus according to the second embodiment. FIG. 10 is a diagram schematically showing a gradation value adjusting process in the visual inspection apparatus according to the third embodiment.

Exemplary embodiments are described in detail below with reference to the drawings. Note that the front-back direction, the up-down direction, and the left-right direction in the following description are directions defined for the sake of convenience in order to simplify the description, and do not limit the actual usage conditions.

(Embodiment 1)
1 and 2 schematically show an appearance inspection apparatus 1 according to the first embodiment. This visual inspection apparatus 1 is, for example, an apparatus for inspecting the defects of a semiconductor component or the like as an inspection object (hereinafter referred to as a work W) from its appearance. The term "defect" as used herein means scratches, dents, adherence of foreign matter, and the like.

A visual inspection apparatus 1 has a box-shaped housing 2 . Inside the housing 2, three tray storage units 4, 5, 6 and one inspection unit 3 are provided. Tray storage units 4, 5, and 6 include a pre-inspection tray storage unit 4 for storing trays in which works W before inspection are arranged, and a non-defective tray storage unit for storing trays in which works W judged to be free of defects are arranged. and a defective product tray storage unit 6 for storing trays in which works W determined to be defective are arranged. Details of the inspection unit 3 will be described later. The arrangement of the three tray storage units 4, 5, 6 and one inspection unit 3 is not particularly limited.

A rail 7 extending along the direction in which the tray storage units 4, 5, 6 and the inspection unit 3 are arranged is provided above the tray storage units 4, 5, 6 and the inspection unit 3. As shown in FIG. An arm 8 for holding a workpiece W is supported on the rail 7 so as to be slidable along the rail 7 . A holding portion 8a that actually holds the workpiece W is provided at the tip (here, the lower end) of the arm 8. As shown in FIG. The holding portion 8a sucks and holds the work W by vacuum suction. The holding portion 8a is configured to be vertically movable. The arm 8 moves along the rail 7 to move the workpiece W held by the holding portion 8 a from the pre-inspection tray storage portion 4 to the inspection portion 3 , the non-defective product tray storage portion 5 or the defective product tray storage portion 6 . transport. The arm 8 is controlled by an arm control section 101 of a controller 100 which will be described later.

The trays stored in the respective tray storage sections 4, 5 and 6 are movable in a direction perpendicular to both the rails 7 and the vertical direction by a conveyor (not shown). This conveyor allows alignment between the arm 8 and the tray. Alternatively, the rail 7 may be movable in a direction perpendicular to both the rail 7 and the vertical direction, and the arm 8 and the tray may be aligned by moving the rail 7 .

The inspection unit 3 has a wall portion 3a in the shape of a rectangular tube. The inspection unit 3 includes one camera 10 that acquires an image of the work W, and a plurality of (here, four) illumination devices 20 that irradiate an inspection area of the work W with light inside the wall 3a.

In this embodiment, only one camera 10 is provided for one inspection unit 3 . That is, the inspection unit 3 of the visual inspection apparatus 1 has a single camera as an image acquisition device. A camera 10 photographs the workpiece W held by the arm 8 . The camera 10 is a camera capable of acquiring a color image of the workpiece W. As shown in FIG. The camera 10 is configured to be able to communicate with the controller 100 and is controlled by an optical system control section 102 of the controller 100 as shown in FIG.

As shown in FIG. 2 , the plurality of lighting devices 20 are arranged so as to surround the camera 10 . Each lighting device 20 is configured to emit light in the visible light region. Specifically, white light, red light, green light, and blue light having different wavelengths are irradiated singly or in combination. Each lighting device 20 is supported by the wall portion 3a. The angle of incidence of the light emitted from each illumination device 20 on the workpiece W and the amount of light are independently adjusted by the optical system control section 102 of the controller 100 . That is, the controller 100 can operate only some of the lighting devices 20 and stop the other lighting devices 20, or operate all the lighting devices 20 and vary the light amount of each lighting device 20. can. The number and arrangement of lighting devices 20 are not particularly limited as long as two or more lighting devices 20 are provided. In particular, the illumination device 20 may be arranged along the entire circumferential direction of the inner peripheral surface of the wall portion 3a so as to surround the entire periphery of the camera 10 .

The visual inspection apparatus 1 includes a display device 30 (see FIG. 3) capable of displaying an image captured by the camera 10 and the results of the visual inspection described later. The display device 30 displays images and the like sent via the controller 100 . The display device 30 may be configured to display icons for operating the camera 10 and the lighting device 20 .

<Control system>
The appearance inspection apparatus 1 is operated and controlled by a controller 100 . The controller 100 has a processor having a CPU, a memory storing a plurality of modules, and the like. The controller 100 has a function of inspecting the workpiece W for defects based on the image captured by the camera 10 . In particular, the controller 100 inspects the workpiece W for defects based on the color image captured by the camera 10 . Such functions are stored as software in memory modules. The number of processors and memories is not limited to one, and the controller 100 may have two or more processors and memories.

As shown in FIG. 3, the controller 100 includes an arm control unit 101 that controls the operation of the arm 8, an optical system control unit 102 that controls the camera 10 and each lighting device 20, and a predetermined an image processing unit 103 that performs processing, an image combining unit 104 that combines images processed by the image processing unit 103 to generate a composite image, a determination unit 105 that determines the presence or absence of a defect based on the composite image, have The arm control unit 101, the optical system control unit 102, the image processing unit 103, the image synthesizing unit 104, and the determination unit 105 are examples of modules stored in the memory.

The arm control section 101 controls the movement of the arm 8, the operation of the holding section 8a, and the like. The arm control unit 101 acquires information about the position of the arm 8 from a position sensor (not shown) provided on the arm 8 . Arm control unit 101 moves arm 8 so that arm 8 is positioned at a desired position based on information from the position sensor. The arm control unit 101 controls the vertical movement and the suction operation of the holding unit 8a so that the workpiece W is properly transported.

The optical system control unit 102 adjusts the focus of the camera 10, and adjusts the position, angle, amount of light, color of light emitted, and the like of each lighting device 20. FIG. The camera 10 acquires a color image of the work W according to a control signal from the optical system control section 102 . In the first embodiment, each lighting device 20 irradiates the work W with white light. The camera 10 acquires a color image of the work W irradiated with white light from each illumination device 20 based on the control signal from the optical system control unit 102 .

The image processing unit 103 extracts RGB images from the color image acquired by the camera 10 . That is, the image processing unit 103 extracts 8-bit monochromatic light images of red, green, and blue from the RGB 24-bit image. A first image and a second image having a different color from the first image are selected from the red, green, and blue 8-bit images generated by the image processing unit 103 .

The image synthesizing unit 104 synthesizes the first image and the second image selected by the image processing unit 103 by adding the gradation values of the plurality of pixels forming each image. The gradation value is, for example, a luminance gradation value. The details of image composition will be described later. Note that the gradation value of saturation may be used as the gradation value.

The determination unit 105 determines whether or not the workpiece W has a defect based on the synthesized image generated by the image synthesis unit 104 . Determination by the determination unit 105 will be described later.

The controller 100 outputs a control signal to the display device 30 to display the determination result of the determination unit 105 . When the determination unit 105 determines that there is a defect, the controller 100 causes the display device 30 to display the defective portion of the work W so that the defect can be identified.

The controller 100 causes the storage unit 40 to store the composite image of the workpiece W determined to be non-defective by the determination unit 105 . The stored composite image is used for determination by the determination unit 105, for example.

<Image processing and image synthesis>
Processing of the image processing unit 103 and the image synthesizing unit 104 will be described.

The image processing unit 103 generates 8-bit monochromatic light images of red, green, and blue from the 24-bit color image acquired by the camera 10, as shown in FIG. At this time, for example, if the defect C is yellow, as shown in FIG. 4, the defect C appears in the red monochromatic light image and the green monochromatic light image, while the defect C does not appear. The image processing unit 103 selects the first image and the second image from these images. In the first embodiment, for example, two images having a large contrast around the defect candidate (difference in luminance between the candidate for defect C and its surroundings) can be selected as the first image and the second image. . In the following description, it is assumed that a red monochromatic light image is selected as the first image and a green monochromatic light image is selected as the second image.

Next, the processing of the image synthesizing unit 104 will be described with reference to FIG. The image synthesizing unit 104 actually synthesizes two-dimensional images, but here, for the sake of simplicity, the description will be based on a one-dimensional profile.

The profiles shown in FIG. 5 are one-dimensional profiles of the first image, the second image, and the composite image including the defect C portion. (a) is the first image, (b) is the second image, and (c) is the composite image. The vertical axis represents luminance gradation values. Since the first and second images are 8-bit images, the minimum value of the gradation values in FIGS. The value is 255.

The image synthesizing unit 104 adds the gradation values of the plurality of pixels forming the first image and the gradation values of the plurality of pixels forming the second image for each corresponding pixel. As a result, the synthesized image has 511 gradations and the maximum gradation value is 510, as shown in FIG. 5(c). In FIG. 5(c), since the vertical axis of the graph is shown in a reduced scale, there appears to be no difference from FIGS. 5(a) and 5(b). and the gradation values of the peripheral portion of the defect C are large. The image synthesizing unit 104 adds the gradation values together without distinguishing colors. For example, when the gradation value of an arbitrary pixel X in the first image (red light image) is 110 and the gradation value of the pixel corresponding to the pixel X in the second image (green light image) is 130 , the gradation value of the pixel corresponding to the pixel X in the synthesized image is 240.

<Defect judgment>
Next, determination by the determination unit 105 will be described. Again, to simplify the explanation, the explanation will be based on the one-dimensional profile.

As shown in FIG. 6, the determination unit 105 extracts a portion whose gradation value is within a predetermined range from the synthesized image generated by the image synthesis unit 104, and determines whether or not there is a defect C based on the extracted portion. do. Specifically, the determination unit 105 extracts an 8-bit range, that is, a range of 256 gradations from the synthesized image. The determination unit 105 can extract, for example, a range from 129 to 384 with 256 interposed therebetween. The range of 256 gradations extracted by the determination unit 105 may be arbitrarily set by the user, or may be a preset range. The range to be extracted may be 7 bits or less or 9 bits or more as long as it is smaller than 16 bits and includes all the gradation values of the defect candidate portion.

Then, the determination unit 105 determines the presence or absence of the defect C based on the extracted image for 256 gradations. A specific mode of determining the defect C is not particularly limited. For example, the determining unit 105 determines that the work W has a defect C when there is a portion where the absolute value of the difference in tone value is greater than or equal to a predetermined value in comparison with the non-defective composite image stored in the storage unit 40. You may Note that the determination unit 105 determines that at least the inspection synthesized image and the comparison synthesized image are adjusted so that the average values of the gradation values of the inspection synthesized image and the comparison synthesized image are the same when comparing with the non-defective synthesized image. One of the gradation values may be uniformly increased or decreased.

<flowchart>
FIG. 7 is a flow chart showing operation processing of visual inspection of a workpiece by the controller 100 .

First, in step S<b>1 , the controller 100 controls the arm 8 to take out the workpiece W before inspection and move it above the inspection section 3 .

Next, in step S<b>2 , the controller 100 irradiates the workpiece W with white light from the illumination device 20 .

Next, in step S3, the controller 100 acquires a color image of the work W irradiated with white light.

Next, in step S4, the controller 100 generates a first image and a second image from the acquired color images. The controller 100 generates a monochromatic light image of each color of RGB from the color image, and selects the first image and the second image from among them to generate the first image and the second image.

Next, in step S5, the controller 100 synthesizes the generated first image and second image. The controller 100 synthesizes the first image and the second image by adding the gradation values of each pixel of each image.

Next, in step S6, the controller 100 extracts a portion having a predetermined range of gradation values from the composite image generated in step S5. Controller 100 extracts an 8-bit range from the composite image.

Then, in step S7, the controller 100 determines whether or not there is a defect C based on the 8-bit image extracted in step S6. If YES, the controller 100 proceeds to step S8. On the other hand, the controller 100 advances to step S9 when the answer is NO that there is a defect C.

In step S8, after notifying that there is no defect C by the display device 30, the process returns and preparations are made for inspecting the next work W. FIG.

On the other hand, in the step S9, the controller 100 causes the display device 30 to display an image showing the location of the defect C to inform that the defect C exists. After that, the controller 100 returns and prepares to inspect the next workpiece W. FIG.

Therefore, according to the first embodiment, the work W is irradiated with light, an image of the work W irradiated with light is acquired, and the first image, which is a monochromatic light image of the first color, is obtained from the acquired image of the work W. and a second image that is a monochromatic light image of a second color different from the first color, and the first image and the second image are obtained by obtaining the respective gradation values of a plurality of pixels constituting each image. are added together to synthesize, and from the resulting synthesized image, a portion having a gradation value within a predetermined range is extracted, and the presence or absence of defect C is determined based on the extracted portion. As a result, since the image is synthesized by adding the gradation values, the difference in the gradation values around the defect becomes large and the contrast becomes large. In addition, by extracting a portion whose gradation value is within a predetermined range, the periphery of the defect can be emphasized. Thereby, the inspection accuracy of the appearance inspection can be improved.

In addition, in the first embodiment, two-color images of RGB are obtained as the first image and the second image from the captured image of the work W. FIG. If it is an image of an object to be inspected illuminated with white light, an image of each color of RGB can be generated from the image. This eliminates the need to obtain an image in which the inspection object is irradiated with each color of RGB, thereby shortening the inspection time.

(Embodiment 2)
Hereinafter, Embodiment 2 will be described in detail with reference to the drawings. In the following description, the same reference numerals are assigned to the same parts as in the first embodiment, and detailed description thereof will be omitted.

In the second embodiment, processing in the image processing unit 103 and the image synthesizing unit 104 is different from that in the first embodiment. Specifically, the image processing unit 103 generates a monochromatic light image of each color of RGB from the color image, and then, among the monochromatic light images, converts the monochromatic light image of the color with the highest contrast around the defect candidate into the first image. , and the monochromatic light image of the color with the smallest contrast around the defect candidate is taken as the second image. Next, as shown in FIG. 8, the image processing unit 103 inverts the gradation values of the second image and executes a reversed image obtaining step of obtaining a reversed image. The image processing unit 103 inverts the gradation value based on the gradation value 127.5, which is the median value of the gradation, so that the gradation value 0 becomes 255 and the gradation value 255 becomes 0.

As shown in FIG. 9, the image synthesizing unit 104 synthesizes the first image (FIG. 9(a)) and the reverse image (FIG. 9(b)) to generate a synthetic image (FIG. 9(c)). do. The image synthesizing unit 104 synthesizes the first image and the reverse image by adding the gradation values of the pixels of the first image and the reverse image, as in the first embodiment. In this way, by adding the inverted image of the image with low contrast around the defect candidate to the image with high contrast around the defect candidate, the gradation value of the background portion is reduced as shown in FIG. It can be made as small as possible, and the contrast around the defect can be emphasized.

Then, the determination unit 105 extracts a range of 256 gradations from the composite image, as in the first embodiment. The determination unit 105 determines whether or not there is a defect C based on the extracted image.

Therefore, according to the second embodiment, the inverted image obtained by inverting the gradation value of the monochromatic light image of the color with the lowest contrast around the defect candidate is the monochromatic light image of the color with the highest contrast around the defect candidate. By synthesizing with one image, the gradation value of the background portion can be reduced as much as possible without substantially affecting the gradation value of the defective portion. As a result, the contrast of the defect portion can be enhanced, and the inspection accuracy of the appearance inspection can be further improved.

(Embodiment 3)
Hereinafter, Embodiment 3 will be described in detail with reference to the drawings. In the following description, the parts common to those of the first and second embodiments are denoted by the same reference numerals, and detailed description thereof will be omitted.

In the third embodiment, processing in the image processing unit 103 and the image synthesizing unit 104 is different from those in the first and second embodiments. Specifically, the image processing unit 103 generates a monochromatic light image of each color of RGB from the color image, selects the first image and the second image, and averages the gradation value of each pixel of the first image. A step for correcting the tone value of each pixel of the first image and the tone value of each pixel of the second image so that the tone value of each pixel of the first image and the average value of the tone values of each pixel of the second image are the same. Execute the adjustment process. For example, as shown in FIG. 10, the image processing unit 103 calculates the average values of the gradation values of the first image and the second image, and adjusts the average values so that the average values are the same. , correction is performed to decrease the gradation value of each pixel of the first image, and correction is performed to increase the gradation value of each pixel of the second image with respect to the second image.

The image synthesizing unit 104 synthesizes the corrected first image and the corrected second image to generate a synthesized image. The image synthesizing unit 104 synthesizes the two images by adding the gradation values of the pixels of the corrected first image and the corrected second image, as in the first embodiment.

Then, the determination unit 105 extracts a range of 256 gradations from the composite image, as in the first embodiment. The determination unit 105 determines whether or not there is a defect C based on the extracted image.

Therefore, according to the third embodiment, by aligning the average brightness of the first image and the second image, when the light receiving sensitivity of the camera 10 is different between RGB, or when the wavelength balance of the RGB of the lighting device 20 is different. Even in such cases, these effects can be reduced. Thereby, the inspection accuracy of the appearance inspection can be further improved.

(Other embodiments)
The technology disclosed herein is not limited to the above-described embodiments, and substitutions are possible without departing from the scope of the claims.

For example, in the above-described embodiment, the work W is irradiated with white light, a color image is acquired by the camera 10, and a monochromatic light image of each color of RGB is acquired from the color image. Not limited to this, the lighting device 20 irradiates the work W with red light to obtain a red monochromatic light image, irradiates green light to obtain a green monochromatic light image, and irradiates blue light to obtain a blue light image. of monochromatic light images may be acquired. In this case, it is necessary to acquire monochromatic light images of respective colors with the photographing position of the camera 10 fixed, that is, with the coordinates of the workpiece W being the same in each image.

Further, in the above-described embodiment, the image processing unit 103 selects the first image and the second image by generating a monochromatic light image of each color of RGB. Not limited to this, for example, a table or the like indicating which monochromatic light image is to be generated for each workpiece W is stored in advance in the memory of the controller 100, and the image processing unit 103 refers to the table stored in the memory. , may generate only a monochromatic light image of a preset color.

Also, the second embodiment and the third embodiment described above may be combined. That is, after making the average value of the gradation values of the first image and the second image the same, the gradation values of the second image may be inverted to create an inverted image. As a result, when the gradation value of the first image and the gradation value of the reverse image are added, the gradation value of the background portion becomes smaller than that of the defect portion, and the contrast around the defect is emphasized as much as possible. can be done.

Further, in Embodiments 1 to 3 described above, a monochromatic light image of each color of RGB is generated, but a monochromatic light image of each color of YCM may be generated.

Further, in Embodiments 1 to 3 described above, monochromatic light images of two colors are synthesized, but monochromatic light images of three or more colors may be synthesized. Also, a plurality of white light images with different lighting conditions (incident angle and light amount of light to the workpiece W) of the lighting device 20 may be synthesized.

Further, although the two-dimensional image of the work W is used in the first to third embodiments described above, a three-dimensional image may be used. Even in this case, each image is photographed so that the coordinates of the work W are the same.

The above-described embodiments are merely examples, and should not be construed as limiting the scope of the present disclosure. The scope of the present disclosure is defined by the claims, and all modifications and changes within the equivalent range of the claims are within the scope of the present disclosure.

The technology disclosed herein is useful when inspecting defects formed on the surface of an inspection object.

1 Appearance inspection device 10 Camera 20 Lighting device 100 Controller (inspection unit)
W work (object to be inspected)

Claims (7)

A visual inspection method for inspecting defects formed on the surface of an inspection object,
an irradiation step of irradiating the inspection object with light;
an imaging step of acquiring an image of the inspection object irradiated with light;
Monochromatic acquiring a first image that is a monochromatic light image of a first color and a second image that is a monochromatic light image of a second color different from the first color from the acquired image of the inspection object. an optical image acquisition step;
an image synthesizing step of synthesizing an image based on the first image and an image based on the second image by adding up the tone values of the plurality of pixels constituting each image for each corresponding pixel;
A determination step of determining the presence or absence of the defect based on the synthesized image obtained in the image synthesis step;
including
The determining step is a step of extracting a portion having a gradation value within a predetermined range from the synthesized image obtained in the image synthesizing step, and determining the presence or absence of the defect based on the extracted portion. Characteristic appearance inspection method. In the appearance inspection method according to claim 1,
The light irradiated in the irradiation step is white light,
The appearance inspection, wherein the monochromatic light image acquiring step is a step of acquiring two color images of RGB from the image acquired in the imaging step as the first image and the second image. Method. In the appearance inspection method according to claim 2,
In the monochromatic light image acquisition step, a monochromatic light image of a color having the highest contrast around the defect candidate among RGB is used as the first image, and a monochromatic light image of a color having the smallest contrast around the defect candidate is used as the first image. A step of forming a second image,
Further comprising a reversed image obtaining step of obtaining a reversed image by inverting the gradation value in the second image before the image synthesizing step,
The visual inspection method, wherein the image synthesizing step is a step of synthesizing the first image and the reverse image. In the appearance inspection method according to claim 1 or 2,
Before the image synthesizing step, the first image is generated so that the average value of the gradation values of the pixels of the first image and the average value of the gradation values of the pixels of the second image are the same. further comprising a gradation value adjusting step of correcting the gradation value of each pixel of the second image and the gradation value of each pixel of the second image;
The image synthesizing step is a step of synthesizing the first image after adjusting the tone value of each pixel and the second image after adjusting the tone value of each pixel. Appearance inspection method. A visual inspection apparatus for inspecting defects formed on the surface of an inspection object,
an irradiation unit that irradiates the inspection object with light;
an imaging unit that acquires an image of the inspection object irradiated with light;
an inspection unit that inspects the presence or absence of defects from the acquired image of the inspection object,
The inspection unit
A first image that is a monochromatic light image of a first color and a second image that is a monochromatic light image of a second color different from the first color from the image of the inspection object acquired by the imaging unit; and get
generating a composite image by synthesizing an image based on the first image and an image based on the second image by adding up the tone values of the plurality of pixels constituting each image for each corresponding pixel;
A visual inspection apparatus, wherein a portion having a gradation value within a predetermined range is extracted from the composite image, and the presence or absence of the defect is determined based on the extracted portion. In the appearance inspection apparatus according to claim 5,
The light emitted by the irradiation unit is white light,
The visual inspection apparatus, wherein the inspection unit acquires images of two colors of RGB as the first image and the second image from the images acquired by the imaging unit. A visual inspection method for inspecting defects formed on the surface of an inspection object,
an irradiation step of irradiating the inspection object with light;
an imaging step of acquiring one or more images of the inspected object irradiated with light;
When there is one image, each gradation value of a plurality of pixels constituting each monochromatic light image is added for each corresponding pixel to a plurality of monochromatic light images obtained from the one image. Each monochromatic light image is synthesized by combining them, and when there are a plurality of said images, each image is synthesized by adding each gradation value of a plurality of pixels constituting each image for each corresponding pixel. an image synthesizing process;
A determination step of determining the presence or absence of the defect based on the synthesized image obtained in the image synthesis step;
including
The determining step is a step of extracting a portion having a gradation value within a predetermined range from the synthesized image obtained in the image synthesizing step, and determining the presence or absence of the defect based on the extracted portion. Characteristic appearance inspection method.

Images