JP2023125709A - Image inspection system and image inspection method - Google Patents

Abstract

To provide an image inspection system capable of shortening an inspection time required for inspection while suppressing deterioration in the accuracy of inspection.SOLUTION: An image inspection system 100 for inspecting an object using multi-wavelength images comprises: an imaging section that acquires a confirmation image by receiving light from a first object via a spectroscopic section for dispersing the light; and a processing section that performs processing on the confirmation image acquired by the imaging section. The processing section generates each image data for respective wavelengths constituting multi-wavelengths included in the confirmation image on the basis of the confirmation image and spectroscopic characteristics of the spectroscopic section, designates a plurality of first wavelengths included in the multi-wavelengths, calculates a correct answer rate of first combined image data on the basis of the first combined image data obtained by combining each image data in the designated first wavelengths and a predetermined correct answer image corresponding to the first object, and determines a plurality of second wavelengths used for inspecting an inspection image obtained by imaging a second object by the imaging section as the plurality of designated first wavelengths if the correct answer rate is equal to or greater than a threshold value.SELECTED DRAWING: Figure 9A

Description

The present disclosure relates to an image inspection system and an image inspection method that inspect an object using multi-wavelength images.

Utilizing spectral information from a large number of bands, each narrow band, for example, several dozen bands, enables detailed analysis of objects that is not possible with conventional RGB images. A camera that acquires information at multiple wavelengths is called a "hyperspectral camera." Hyperspectral cameras are used in various fields such as food inspection, biological testing, drug development, and mineral component analysis.

International Publication No. 2021/192891

When inspecting an object to be inspected using images captured by a hyperspectral camera, inspection accuracy can be improved by increasing the number of multiple wavelengths included in the image of the inspection object, but the inspection time becomes longer. On the other hand, when the number of wavelengths is small, the inspection time is shortened, but the inspection accuracy is reduced.

The present disclosure provides an image inspection system and an image inspection method that can shorten the inspection time required for inspection while suppressing a decrease in inspection accuracy.

One aspect of the present disclosure is an image inspection system that inspects an object using a multi-wavelength image, which receives light from a first object through a spectroscope that separates light to produce a confirmation image. and a processing section that processes the confirmation image obtained by the imaging section, and the processing section is configured to perform processing on the confirmation image and the spectral characteristics of the spectroscopic section. Generate each image data of each wavelength that constitutes the multiple wavelengths included in the confirmation image, specify a plurality of first wavelengths included in the multiple wavelengths, and generate a first image data that combines each image data of the specified first wavelengths. A correct answer rate of the first combined image data is calculated based on the first combined image data and predetermined correct answer image data corresponding to the first object, and if the correct answer rate is equal to or higher than a threshold, the The image inspection system determines a plurality of second wavelengths to be used for inspection of an inspection image obtained by capturing an image of a second object by an imaging unit, as the plurality of designated first wavelengths.

One aspect of the present disclosure is an image inspection method for inspecting a target object using a multi-wavelength image, the image inspection method includes receiving light from a first target object through a spectroscopic unit that splits light to obtain a confirmation image. and a processing step of performing processing on the obtained confirmation image, and the processing step includes an image acquisition step of obtaining the confirmation image based on the confirmation image and the spectral characteristics of the spectroscopic section. a step of generating each image data of each wavelength constituting the multiple wavelengths included in the multiple wavelengths, a step of specifying a plurality of first wavelengths included in the multiple wavelengths, and a step of combining each image data of the specified first wavelengths. calculating a correct answer rate of the first combined image data based on the first combined image data and a predetermined correct answer image corresponding to the first object, and the correct answer rate is equal to or higher than a threshold value. In this case, an image inspection method includes the step of determining a plurality of second wavelengths to be used for inspecting an inspection image obtained by imaging a second object to be the plurality of specified first wavelengths. .

According to the present disclosure, the test time required for the test can be shortened while suppressing a decrease in test accuracy.

Block diagram of an image inspection system according to an embodiment of the present disclosure Diagram showing an example of a confirmation image acquired by the camera Diagram showing an example of each image data of each wavelength included in the confirmation image Conceptual diagram of correct answer image and confirmation image Diagram showing an example of the specified area in the correct answer image Graph showing the amount of light received for each wavelength in the specified area Graph showing the contribution of each wavelength to the number A diagram showing an example of a state in which image data for each wavelength included in a confirmation image is arranged in order of contribution. A graph showing an example of the correct answer rate for combined image data that combines image data of various combinations of first wavelengths. Diagram showing an operation example for manually selecting image data for each wavelength included in the confirmation image Flowchart showing an example of the procedure of the image inspection method executed by the image inspection system Flowchart showing an example of the procedure of the image inspection method executed by the image inspection system (continued from FIG. 8A) A graph showing an example of the correct answer rate for the number of wavelengths when wavelengths are added and combined in order of their contribution. A graph showing an example of the correct answer rate for the number of wavelengths when wavelengths are arbitrarily combined without depending on the degree of contribution Block diagram of modification 1 of the image inspection system Block diagram of modification 2 of the image inspection system Block diagram of modification 3 of image inspection system

Hereinafter, embodiments will be described in detail with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of well-known matters or redundant explanations of substantially the same configurations may be omitted. This is to avoid unnecessary redundancy in the following description and to facilitate understanding by those skilled in the art. The accompanying drawings and the following description are provided to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter recited in the claims.

FIG. 1 is a block diagram of an image inspection system according to an embodiment of the present disclosure. The image inspection system 100 includes a camera 10, a lighting device 20, a control device 30, and a display operation device 40. The image inspection system 100 inspects the characteristics of the imaged object (workpiece, subject) by analyzing each image data of each wavelength that constitutes the multiple wavelengths included in the image. do. In this embodiment, the camera 10, the illumination device 20, the control device 30, and the display operation device 40 are each configured as independent devices. For example, the lighting device 20 may be detachable from the camera 10. The control device 30 is connected to the camera 10, the lighting device 20, and the display operation device 40 via an interface (wired or wireless) not shown. That is, the camera 10, the lighting device 20, the control device 30, and the display operation device 40 have communication devices.

The camera 10 is a device capable of capturing an image of a target object that includes multi-wavelength information such as spectral information of several tens of bands, that is, a so-called multi-wavelength image. A multi-wavelength image is an image that includes components of a large number of wavelengths (for example, three or more, for example, 20 wavelengths (20 bands)). Camera 10 is also called a hyperspectral camera. The camera 10 is installed at a fixed position, for example, and images an object. The camera 10 includes a spectroscopic section 11, an imaging section 12, and a storage section 13.

The spectroscopic section 11 may be configured with a transmission type (for example, a filter or prism) optical member, or a reflection type (for example, a diffraction grating) optical member. The spectroscopic unit 11 receives light from an object irradiated with light by the illumination device 20, and separates the light into a plurality of wavelengths. That is, the spectroscopic section 11 is arranged corresponding to the imaging section 12 and has a wavelength selection function for each pixel of the imaging section 12. The spectroscopic section 11 may be divided into a plurality of spectral regions. Each spectral region passes a predetermined wavelength. The one or more wavelengths passed in each spectral region may or may not overlap at least in part.

The imaging unit 12 is configured with an imaging device or the like that converts the light received through the spectroscopic unit 11 into an electrical signal and captures an image of the object. The wavelength ranges of each pixel of the imaging unit 12 may or may not overlap at least in part.

The storage unit 13 is composed of a memory, a disk, etc. that stores settings of the camera 10, operating programs, captured images, and the like.

The illumination device 20 is a device that irradiates an object with illumination light and enables the camera 10 to capture an image of the object. The lighting device 20 includes a light projection section 21 and a storage section 22. The light projecting unit 21 is a light source that can irradiate illumination light onto a target object, and is composed of a light emitting element, a light bulb, or the like. The storage unit 22 is composed of a memory, a disk, etc. that stores settings, operating programs, etc. of the lighting device 20. The illumination light from the illumination device 20 may be white light or light having at least one predetermined wavelength.

The control device 30 is a device that controls the image inspection system 100, and is, for example, a PC (Personal Computer). The control device 30 constitutes a processing unit that performs various processes on images captured by the camera 10. The control device 30 includes a control section 31, a calculation section 32, a determination section 33, and a storage section 34. The control unit 31, the calculation unit 32, and the determination unit 33 read an image inspection program stored in the storage unit 34, for example, and control the operation of the control device 30 and, in turn, the image inspection system 100.

The control unit 31 is composed of a processor or the like that controls the overall operation of the control device 30. Further, the control unit 31 performs, for example, instructions for controlling the camera 10 and the lighting device 20 and various settings (eg, camera settings, lighting settings).

The arithmetic unit 32 is composed of, for example, a processor specialized in arithmetic processing. The calculation unit 32 performs, for example, deriving the correct answer rate and contribution degree, which will be described later, generating various images, and specifying the wavelength to be inspected among the wavelengths included in the images.

The determination unit 33 is composed of a processor or the like specialized for various determination processes. The determination process may include, for example, determining whether the correct answer rate of the confirmation image, inspection image, or combined image data, which will be described later, satisfies predetermined requirements. The confirmation image is an image captured by the camera 10 and includes components of all wavelengths, and is an image used for inspection preparation (designation of wavelengths used for inspection). The inspection image is an image captured by the camera 10 and includes components of all wavelengths, and is an image used for inspection. The combined image data is data corresponding to an image that includes components of some of the wavelengths included in the confirmation image or the inspection image. Note that the confirmation image may be used for the inspection.

The storage unit 34 is configured as a memory, a disk, or the like that stores settings, operating programs, etc. of the control device 30.

The display operation device 40 functions as a display device that displays images captured by the camera 10, processing results by the control device 30, and the like, and also functions as an operation unit through which the user inputs various operations. The display operation device 40 includes an initial setting section 41, an inspection condition setting section 42, an image display section 43, and a result display section 44.

The initial setting section 41 is one of the operation sections that accepts operation input for initial setting of the display operation device 40 by the user, and is constituted by an input device such as a keyboard or a mouse. The initial setting section 41 performs initial settings based on this operation input. The inspection condition setting unit 42 is one of the operation units that accepts operation input of inspection conditions of the image inspection system 100 by the user, and is configured by an input device such as a keyboard or a mouse. The inspection condition setting section 42 sets inspection conditions based on this operational input. The inspection conditions include, for example, the number of wavelengths used for image inspection, the correct answer rate used for image inspection, or other inspection conditions. The correct answer rate is an index for identifying whether the object to be inspected is a good product or a defective product. One input device can also serve as the operation section for the initial setting section 41 and the inspection condition setting section 42. The set initial conditions and test conditions may be sent to the control device 30 and held in the storage unit 34.

The image display unit 43 is a display that displays images captured by the camera 10 and the like. The result display unit 44 is a display that displays the results of the image processing performed by the control device 30. One display can also serve as the image display section 43 and the result display section 44.

Further, the initial setting section 41 and the inspection condition setting section 42 may be configured by an input device, and the image display section 43 and the result display section 44 may be configured by a display separate from the input device.

FIG. 2A is a diagram showing an example of a confirmation image GC that is an image acquired by the camera 10. FIG. 2B is a diagram showing each image data gd of each wavelength included in the confirmation image GC of FIG. 2A.

The illumination device 20 irradiates illumination light onto a first object to be inspected by the image inspection system 100 . The spectroscopic unit 11 receives light from an object irradiated with light by the illumination device 20, and separates the light into a plurality of wavelengths. The imaging section 12 converts the light received via the spectroscopic section 11 into an electrical signal, and obtains the confirmation image GC of FIG. 2A. Therefore, the confirmation image GC is an image in which the first target object is reflected.

Similarly, although not shown, the illumination device 20 irradiates the first object to be inspected by the image inspection system 100 with illumination light. The spectroscopic unit 11 receives light from an object irradiated with light by the illumination device 20, and separates the light into a plurality of wavelengths. The imaging section 12 converts the light received via the spectroscopic section 11 into an electrical signal, and obtains the confirmation image GC of FIG. 2A. Therefore, the confirmation image GC is an image in which the first target object is reflected.

In this embodiment, the target object (for example, a first target object or a second target object to be described later) is, for example, a painted object, and can be an inspection target. The first object is used in advance for inspection of the second object using multi-wavelength images. A confirmation image is obtained by imaging the first object. The second object is the actual object to be inspected. An inspection image is obtained by imaging the second object. The first object and the second object may be objects of the same type (objects with similar coatings, objects with similar color distribution, predetermined parts, or completed objects). Note that the first object may also be an actual inspection object, or the first object may be one of the second objects. That is, the same plurality of wavelengths used in the inspection of the object may be used in the subsequent inspection of the object. For example, the first object may be the first object to be inspected among a plurality of second objects.

The control device 30 acquires the confirmation image GC of FIG. 2A from the camera 10. Due to the spectral characteristics of the spectroscopic unit 11, the confirmation image GC includes images of each wavelength constituting multiple wavelengths, such as spectral information of several tens of bands, for example. Based on the confirmation image GC and the spectral characteristics of the spectroscopic section 11, the calculation unit 32 of the control device 30 calculates each image data of wavelength 1 to wavelength 20 constituting the multiple wavelengths included in the confirmation image GC, as shown in FIG. 2B. Generate gd. Note that in each image data gd, the greater the amount of light received at the wavelength indicated by the predetermined image data gd, the larger (whiter) the pixel value of this predetermined image data gd becomes; The pixel value of the predetermined image data gd becomes smaller (becomes black). In this example, the number of wavelengths in the multiple wavelengths is 20, but the number of wavelengths is not limited to a specific number, and may be 19 or less or 21 or more.

FIG. 3 is a conceptual diagram of the correct answer image GA and the confirmation image GC. The correct answer image GA is, for example, an image of the correct reference object itself, acquired in advance by the control device 30 and stored in the storage unit 34, and specifically, a sample image serving as a color sample of the object. In other words, the correct image GA is an image that shows the color distribution targeted by the confirmation image or test image. The correct answer image GA is an image of the first object, similar to the confirmation image GC. Moreover, the correct answer image GA is a confirmation image GC. The determination unit 33 of the control device 30 determines whether or not the confirmation image GC obtained by imaging the first object to be inspected with the camera 10 matches the correct answer image GA. It is determined whether the item matches the reference object and is a correct item (for example, whether there is a defect).

The concept used in this determination is the correct answer rate. The determination unit 33 of the control device 30 compares the amount of light received in the image of each wavelength of the multiple wavelengths included in the confirmation image GC with the amount of light received in the image of each wavelength in the correct answer image GA, and determines the color of the confirmation image GC ( A portion where the color (distribution of the amount of received light of each wavelength) differs from the color of the correct image GA (distribution of the amount of received light of each wavelength) is determined. Then, the calculation unit 32 calculates the area of the portion determined to be different, the so-called erroneous determination area. Further, the calculation unit 32 calculates the proportion of the area of this misjudgment to the area of the object. The correct answer rate is determined by this series of calculations, and the formula for the correct answer rate is, for example, (correct answer rate=1−ratio of area of false judgment to area of target object).

In the example of FIG. 3, the correct answer image GA and the confirmation image GC have an area of 1 ² cm ² . The correct answer image GA has, in its central portion, an area of a second color that is different in color from the surrounding first color, and the area of the area is 0.3 ² cm ² . On the other hand, the confirmation image GC, like the correct answer image GA, has a second color area in its center that is different from the surrounding first color, but the area is ^0.2 cm2 ^. Yes, and the size of the second color area is different from the correct answer image GA. In this case, the correct answer rate of the confirmation image GC is determined as follows.

・Area of object: 1 ² cm ²
・Misjudgment area: 0.3 ² -0.2 ² cm ²
・Correct answer rate: 1-(0.3 ² -0.2 ² )/1 ² =95%

However, the method for calculating the correct answer rate described above is just an example, and the method for calculating the correct answer rate is not limited to the one described here.

FIG. 4A, FIG. 4B, and FIG. 4C are diagrams showing the process of determining the degree of contribution of each wavelength. The degree of contribution is a value indicating the degree to which each wavelength contributes among the multiple wavelengths included in the confirmation image when calculating the correct answer rate of the confirmation image explained in FIG. The degree of contribution is calculated based on the variation in the amount of light received for each wavelength obtained when the first object is imaged by the camera 10 according to each region of the first object. In other words, for each of the wavelengths that make up the multiple wavelengths, there are variations in the amount of light received in each region of the first object (even for one wavelength, the amount of light received differs in different regions), and the variation It can be said that the larger the value, the easier it is to correctly distinguish and identify areas, and the greater the contribution to the correct answer rate.

FIG. 4A shows a region specified in the confirmation image or correct answer image when calculating the degree of contribution. Regions a, b, c, and d in FIG. 4A are regions used in the confirmation image or correct answer image obtained by imaging the first object when determining the degree of contribution of each wavelength. Before acquiring the confirmation image, the calculation unit 32 of the control device 30 may designate areas a, b, c, and d in the confirmation image or the correct answer image and store them in the storage unit 34. Areas a, b, c, and d may be arbitrary areas, for example, areas that are frequently inspected. Further, the display operation device 40 receives an operation input from the user via the inspection condition setting unit 42 to enclose an arbitrary region in the confirmation image or the correct answer image, specifies regions a, b, c, and d, and specifies the regions a, b, c, and d. The area designation information may be transmitted to the control device 30. Then, the calculation unit 32 of the control device 30 may designate the regions a, b, c, and d based on the region designation information from the display operation device 40. The designation of these areas corresponds to the designation of the area in which the correct answer rate of the confirmation image, test image, or combined image data will be calculated later. Further, the designated area may match the area to be inspected in an inspection of an actual inspection image. Furthermore, areas a, b, c, and d may be collectively designated as one area.

FIG. 4B shows a graph showing the amount of received light (distribution of the amount of received light) with respect to the wavelength for each designated area. In FIG. 4B, the amount of light received in each region a, b, c, and d is different for each wavelength shown on the horizontal axis, and the variation in the amount of light received in regions a, b, c, and d is different for each wavelength. For example, at a wavelength around 480 nm, the amount of light received is larger in all regions than at a wavelength longer than 550 nm, but the amount of light received in each region a, b, c, and d also varies greatly. Therefore, it can be said that the above contribution rate is greater for wavelengths around 480 nm than for wavelengths longer than 550 nm. Also, the wavelength

Specifically, the calculation unit 32 of the control device 30 compares the amounts of received light in the designated areas a, b, c, and d, and calculates the degree of contribution of each wavelength. In calculating the degree of contribution, for example, multivariate analysis (eg, principal component analysis) may be used, or Mahalanobis distance may be used. The Mahalanobis distance is an index that indicates how far a group of values is from the average (center of gravity) in standard deviation units, and is also a measure that indicates how similar two data groups are. Data groups with high similarity have a high overlap rate and the Mahalanobis distance is short. At this time, the part where there is no overlap between the two data groups becomes the difference between the two data groups. For example, the shorter the Mahalanobis distance, the lower the contribution, and the longer the Mahalanobis distance, the higher the contribution.

FIG. 4C shows a graph showing the contribution of selected wavelengths to number. In FIG. 4C, the horizontal axis shows the wavelength number, and the vertical axis shows the degree of contribution. In FIG. 4C, wavelength numbers are arranged in descending order of contribution. That is, in FIG. 4C, when the contribution of each wavelength calculated by the calculation unit 32 is arranged in descending order, wavelength 3, wavelength 7, wavelength 8, wavelength 4, wavelength 5, . . . , wavelength 19, wavelength 20, wavelength 18, and shows changes in contribution at each of these wavelengths.

Note that the areas a, b, c, and d to be inspected in the confirmation image do not have to be designated. That is, the inspection target in the confirmation image may be the entire confirmation image. Multiple images may match the correct answer.

FIG. 5 shows a state in which image data for each wavelength included in the confirmation image is arranged in order of contribution. The arrangement of image data based on the degree of contribution may be performed by the calculation unit 32. The calculation unit 32 arranges the image data for each wavelength included in the confirmation image in the order of contribution, as shown in FIG. 5, according to the contribution determined in the above-described process. This order shown in FIG. 5 is the same as the order of wavelength numbers on the horizontal axis in FIG. 4C. In the control device 30, the calculation unit 32 instructs the display and operation device 40 to display via, for example, a communication unit (not shown), and the image display unit 43 of the display and operation device 40 displays the arranged image data (wavelength numbers: Wavelength 3, wavelength 7, wavelength 8, wavelength 4, wavelength 5, . . . wavelength 19, wavelength 20, wavelength 18) may be displayed.

The calculation unit 32 of the control device 30 specifies a plurality of wavelengths (first wavelengths) from among the multiple wavelengths included in the confirmation image, as in the area surrounded by squares in FIGS. 4B and 4C. In this example, the calculation unit 32 specifies the first wavelength of 10 bands from a total of 20 bands of multiple wavelengths. The calculation unit 32 may designate a plurality of first wavelengths, for example, by combining a plurality of wavelengths in descending order of contribution. The calculation unit 32 may specify the number of wavelengths specified as the first wavelength. In this case, the number of wavelengths may be specified based on a threshold value to be described later, may be predetermined and stored in the storage unit 34, or may be manually specified via the inspection condition setting unit 42 of the display operation device 40. may be specified.

For example, when the first wavelength of 10 bands is specified, the calculation unit 32 may combine the image data for the specified 10 bands to generate the first combined image data. . The first combined image data is compared with the correct answer image and becomes data used for deriving the correct answer rate in the case of multiple wavelengths. The first combined image data is data in which some wavelength components are omitted when compared with the confirmation image in which all wavelengths are taken into account.

Further, the calculation unit 32 calculates the correct answer rate of the generated first combination image data based on the correct answer image corresponding to the first target object. The correct answer rate can be calculated by the method described in FIG. 3. In this case, the first combined image data is used instead of the confirmation image in FIG. 3. In other words, when comparing the confirmation image and the correct answer image, the correct answer rate is calculated taking into account each of all the wavelengths included in the confirmation image, and when comparing the first combined image data and the correct answer image, the The correct answer rate is calculated taking into account the specific wavelength (first wavelength) of the part.

FIG. 6 is a graph showing the correct answer rate for first combined image data obtained by combining image data of various first wavelengths.

The calculation unit 32 combines the plurality of wavelengths in the order of the contribution of each wavelength determined in the process described in FIGS. 4A to 4C, that is, in the order of the wavelengths with the highest contribution in the confirmation image, that is, the plurality of first Image data of wavelengths are specified, and the specified image data are combined to generate first combined image data. In FIG. 6, first combined image data is generated by sequentially increasing the number of wavelengths to be combined in descending order of contribution, and the correct answer rate is plotted on the vertical axis. In this example, the correct answer rate for the image data of wavelength 3, which has the highest contribution, is plotted on the left end of the horizontal axis, and then the image data of wavelengths are plotted in order of contribution, such as wavelength 7, wavelength 8, etc. The correct answer rate of the first combined image data is plotted. At the right end of the horizontal axis, the correct answer rate of the first combined image data, which is a combination of image data of all wavelengths (20 bands), is plotted. Note that the image corresponding to the first combined image data, which is a combination of image data of all wavelengths, is equal to the confirmation image.

By increasing the number of wavelengths (that is, the number of image data of each wavelength) for generating the first combined image data, the correct answer rate of the first combined image data improves. However, as the number of wavelengths to be combined increases, although the correct answer rate of the first combined image data improves, the improvement in the correct answer rate eventually becomes saturated. In the graph of FIG. 6, for example, after the correct answer rate of the first combined image data based on 10 bands of image data reaches a predetermined threshold th, even if more image data is added, the correct answer rate significantly improves. is not expected. Therefore, the calculation unit 32 specifies the threshold th with which the correct answer rate is compared, as shown in FIG. Then, the calculation unit 32 generates the first combination image data, for example, using the image data of the minimum number of wavelengths required to reach the threshold th (the minimum number of bands, for example, the number of wavelengths is 10). The threshold value th may be, for example, a value corresponding to a saturated state of the correct answer rate.

FIG. 7 shows an operation for manually selecting image data for each wavelength included in the confirmation image. For example, image data may be displayed in order of contribution, and operations related to selection may be performed.

In the examples shown in FIGS. 4A to 6, the calculation unit 32 of the control device 30 automatically arranges the multi-wavelength image data included in the confirmation image (for example, in order of contribution) regardless of the user's intention. ) and specifying a plurality of first wavelengths. In this case, the initial setting section 41 of the display and operation device 40 may set the "automatic mode" by accepting a user's operation from, for example, an operation section. The automatic mode is one of the operation modes of the control device 30, and is a mode in which multiple wavelengths (first wavelengths) to be inspected are automatically specified.

On the other hand, in the example of FIG. 7, the initial setting unit 41 of the display operation device 40 receives a user's operation through the operation unit and sets the “manual mode”, so that the image display unit 43 configures multiple wavelengths. Each image data of each wavelength may be displayed together with a check box. The manual mode is one of the operation modes of the control device 30, and is a mode in which a plurality of wavelengths (first wavelengths) to be inspected are manually specified. The inspection condition setting unit 42 specifies arbitrary image data from among the displayed image data in response to the user's operation on the operation unit, and specifies the first wavelength corresponding to the specified image data. good. In this example, the inspection condition setting unit 42 specifies a plurality of first wavelengths by checking the checkbox corresponding to the wavelength (image data) in response to the user's operation on the operation unit, and inputs the wavelength specification information. may be transmitted to the control device 30. Then, the calculation unit 32 of the control device 30 may designate a plurality of first wavelengths based on the wavelength designation information from the display and operation device 40. Thereby, the calculation unit 32 also specifies the number of wavelengths of the first wavelength.

The image display section 43 displays each image data of each wavelength, and may also display identification information (for example, wavelength number) of each wavelength, contribution degree of each wavelength, and the like. Thereby, the user may specify the wavelength (image data) via the operation section of the display and operation device 40 while checking the degree of contribution of each wavelength. In addition, when the first combined image data in which each image data corresponding to the specified first wavelength is combined is displayed, it is also possible to consider whether this first combined image data is easy for people to see. good. For example, by specifying a plurality of wavelengths that are close to each other and generating and displaying the first combined image data, it is easy to clearly confirm the gradation of the first object. Therefore, even when inspecting the image of the second object, the second combined image data (combined image data for inspection) similar to the first combined image data is displayed, so that the image of the second object can be inspected. It becomes easier to see the gradation clearly.

Note that the first combined image data may be a combination of image data corresponding to the first wavelength, and may not be the image itself (it may not be possible until the image is generated). Alternatively, it may be an image (integrated image) in which each image data corresponding to the first wavelength is combined (integrated). Similarly, the second combined image data may be a combination of image data corresponding to the second wavelength, and may not be the image itself (it may not be the image itself). ), or an image (integrated image) in which each image data corresponding to the second wavelength is synthesized (integrated).

Next, inspection of the second object using a multi-wavelength image will be explained.

The calculation unit 32 of the control device 30 acquires a confirmation image based on the imaging of the first object, as described above. The calculation unit 32 compares the combined image data for wavelength designation with the correct answer image, and calculates the correct answer rate of the combined image for wavelength designation. If this correct answer rate satisfies a predetermined correct answer rate (if it is greater than or equal to the threshold th), the calculation unit 32 calculates the number of first wavelengths (for example, 10 wavelengths) included in the combined image data for wavelength designation. A specific wavelength (for example, a wavelength with a high degree of contribution) is determined as a plurality of second wavelengths to be used for inspecting an inspection image obtained by capturing an image of the second object with the camera 10.

The calculation unit 32 also acquires an inspection image based on the imaging of a new second object that is the same type as the first object, and generates an inspection image that includes the determined plurality of second wavelength components. Combined image data (second combined image data) is used. The calculation unit 32 compares the test combination image data and the correct answer image, calculates the correct answer rate of the test combination image data, and the determination unit 33 determines whether the second target object is the correct answer based on this correct answer rate. Determine whether the product is good or defective. For example, the determination unit 33 determines that the second object is a good product when the correct answer rate is equal to or higher than the threshold th, and when the correct answer rate is less than the threshold th, the second object is determined to be a good product. It is determined that the product is defective. Note that the method for deriving the correct answer rate during inspection and the method for specifying the area to be inspected may be the same as the method for specifying the wavelength for inspection. In addition, although we have illustrated that the threshold value for determining whether the second object is good or defective is the same as the threshold value that is compared with the correct answer rate for specifying the wavelength described above, these threshold values are different. It's okay.

The calculation unit 32 sequentially images a plurality of second objects, sequentially acquires inspection images, sequentially generates combined image data for inspection, and sequentially calculates the correct answer rate of the combined image data for inspection. good. Then, the determining unit 33 may sequentially determine whether the second object is a good product or a defective product.

Therefore, when inspecting an inspection image, the image inspection system 100 inspects using the specified second wavelength without inspecting all wavelengths. The time required for image inspection of objects can be reduced. Furthermore, even if the number of wavelengths used for the inspection of the inspection image is reduced, the image inspection system 100 uses the confirmation image of the first object to determine that the wavelength for which a correct answer rate equal to or higher than the threshold th can be obtained is the wavelength to be inspected. Since it has been determined, it is possible to suppress a decrease in the inspection accuracy of the inspection image.

8A and 8B are flowcharts illustrating an example of the procedure of an image inspection method executed by the image inspection system 100.

First, the control device 30 acquires a correct image of a reference object to be used as a reference (step S1). For example, when the inspection condition setting unit 42 instructs to acquire a correct answer image via the operation unit, the control unit 31 acquires the correct answer image from a predetermined server via a network (not shown), and 34.

Subsequently, the control unit 31 of the control device 30 causes the camera 10 to image the object, obtains a confirmation image, and stores it in the storage unit 34 (step S2). The calculation unit 32 specifies predetermined areas in the correct answer image and the confirmation image, such as areas a, b, c, and d shown in FIG. 4A (step S3). These areas a to d correspond to areas to be inspected.

Subsequently, the calculation unit 32 of the control device 30 calculates the correct answer rate of the confirmation image with respect to the correct answer image in the designated area, for example, in the manner shown in FIG. 3 (step S4). In this case, the calculation unit 32 may calculate the correct answer rate for each image data of each designated first wavelength. That is, the calculation unit 32 may calculate the correct answer rate for each single wavelength by comparing the image data of a single wavelength and the correct answer image. Furthermore, the calculation unit 32 calculates the degree of contribution to the correct answer rate for each wavelength, for example, in the manner shown in FIGS. 4A to 4C (step S5).

Subsequently, the calculation unit 32 determines whether or not to automatically specify the wavelength to be inspected (step S6). For example, if the initial setting unit 41 has set the automatic mode, the control unit 31 automatically specifies the wavelength to be inspected (step S6; YES). If the initial setting unit 41 has set the mode to manual mode, the control unit 31 manually specifies the wavelength to be inspected (step S6; No).

When automatically specifying wavelengths, the calculation unit 32 specifies a plurality of first wavelengths from among the multiple wavelengths included in the confirmation image in the manner shown in FIGS. 4C to 6 (step S7). In this case, the calculation unit 32 may specify a plurality of first wavelengths according to inspection conditions such as the number of wavelengths and the correct answer rate (that is, the threshold th). Specifically, the calculation unit 32 specifies a predetermined number of first wavelengths in descending order of correct answer rate or contribution degree, or specifies a plurality of first wavelengths whose correct answer rate or contribution degree exceeds a predetermined threshold set in advance. You may also specify a wavelength. The calculation unit 32 calculates the correct answer rate of the first combined image data by combining each image data corresponding to the specified first wavelength and comparing the first combined image data and the correct answer image. (Step S9).

On the other hand, when the wavelength is not automatically designated (when the wavelength is manually designated), the calculation unit 32 manually designates the image data by designating a plurality of first wavelengths in the manner shown in FIG. S8). The calculation unit 32 generates first combined image data by combining each piece of image data corresponding to the specified first wavelength, and compares the first combined image data with the correct answer image to generate the first combined image data. The correct answer rate of the combined image data is calculated (step S9).

Subsequently, the determination unit 33 of the control device 30 determines whether the correct answer rate of the first combined image data calculated in step S9 is equal to or higher than a predetermined threshold th as shown in FIG. S10). If the correct answer rate is smaller than the predetermined threshold th (step S10; NO), the calculation unit 32 manually specifies the image data by specifying the plurality of first wavelengths again, for example in the manner shown in FIG. (Step S8). Then, the calculation unit 32 executes step S9 again, and the determination unit 33 executes step S10 again.

Note that in step S8, the determination unit 33 instructs the display operation device 40 to perform a guide display (recommended display) for specifying the first wavelength, and the image display unit 43 of the display operation device 40 , this guide display may be performed according to instructions from the control device 30. The guide display makes it easier for the user to manually specify in step S8. The guide display is a display that makes it easy to identify what is recommended to be designated as the first wavelength, such as a combination of image data that has a high correct answer rate or a high degree of contribution. For example, the wavelength number of the wavelength recommended for designation may be highlighted (for example, displayed in bold or colored). For example, a frame of image data of a wavelength recommended for designation or a frame of a check box may be highlighted (for example, displayed with a thick frame or displayed with a colored frame).

If the correct answer rate of the first combined image data calculated in step S9 is equal to or higher than the predetermined threshold th as shown in FIG. The plurality of second wavelengths used for inspection of the inspection image obtained by imaging the second object are determined to be the plurality of first wavelengths specified above (step S11). Information on the plurality of determined first wavelengths may be held in the storage unit 34.

After step S11 ends, the process proceeds to FIG. 8B. Similarly to the first object, the camera 10 images the second object to obtain an inspection image and transmits it to the control device 30. The determination unit 33 of the control device 30 acquires an inspection image from the camera 10 (step S12). Then, the determination unit 33 determines the inspection image, that is, determines whether the second object matches the reference object (step S13). In this case, the determination unit 33 generates second combined image data by combining each image data included in the inspection image and corresponding to the specified second wavelength, and combines the second combined image data with the above-mentioned correct answer. The correct answer rate of the second combined image data is calculated by comparing with the image. In other words, the correct answer rate when a plurality of specified second wavelengths are used is calculated. Then, for example, when the correct answer rate is equal to or higher than the threshold th, the determination unit 33 determines that the second object is a good product, and when the correct answer rate is less than the threshold th, the second object The item may be determined to be defective. The determination unit 33 transmits information on the inspection result as to whether the product is good or defective to the display operating device 40 via the communication device of the control device 30, and the result display unit 44 of the display operating device 40 displays Information on the test results may be acquired via the communication device of the operating device 40 and the test results may be displayed. The user (tester) can check and understand the display of the test results.

Therefore, steps S1 to S11 in FIG. 8A are a preparation stage (a stage of specifying the wavelength to be used for the test) for the test of the second object to be tested. Steps S12 and S13 in FIG. 8B are an inspection in which a plurality of first wavelengths specified in the preparation stage are used as a plurality of second wavelengths for inspection, and a second target object is inspected using images of other wavelengths. It is a stage.

FIGS. 9A and 9B are graphs showing an example of the correct answer rate with respect to the number of combined wavelengths. FIG. 9A shows the correct answer rate of the second combined image data with respect to the number of wavelengths when wavelengths are added and combined in descending order of contribution. FIG. 9A shows a graph substantially the same as FIG. 6. FIG. 9B shows the correct answer rate of the second combined image data with respect to the number of wavelengths when wavelengths are arbitrarily combined without depending on the degree of contribution.

In FIG. 9A, wavelengths are added in descending order of contribution, and second combined image data is generated by combining image data of the combined wavelengths. In this case, the control device 30 uses the second combined image data of wavelength combinations of about 10 bands to perform the same operation as the second combined image data of wavelength combinations of all 20 bands in the random case shown in FIG. 9B. It is possible to obtain a certain percentage of correct answers. As a result, even if the image inspection system 100 is the second combined image data obtained by combining wavelengths in descending order of contribution, the second combined image data obtained by combining all wavelengths (that is, the inspection image ), it is possible to maintain a high correct answer rate while suppressing the amount of data to be processed. Therefore, the image inspection system 100 can ultimately reduce the inspection time required for the image inspection while suppressing a decrease in the accuracy of the image inspection for the object.

FIG. 10 is a block diagram of Modification 1 of the image inspection system. In Modification 1, camera unit 10A includes camera 10 and lighting device 20 in FIG. That is, the camera unit 10A is a device that integrally has the functions of the camera 10 and the illumination device 20.

FIG. 11 is a block diagram of a second modification of the image inspection system. In modification 2, the control unit 30A includes the control device 30 and the display operation device 40 in FIG. That is, the control unit 30A is a device that integrally has the functions of the control device 30 and the display operation device 40.

FIG. 12 is a block diagram of modification 3 of the image inspection system. In modification 3, the camera unit 10A includes the camera 10 and the illumination device 20 in FIG. 1, and the control unit 30A includes the control device 30 and the display operation device 40 in FIG. That is, the third modification is a combination of the camera unit 10A of the first modification and the control unit 30A of the second modification.

In this way, according to the image inspection system 100 of the present embodiment, the image inspection system 100 can shorten the inspection time while suppressing a decrease in the inspection accuracy of the inspection target using multi-wavelength images. In addition, by increasing the wavelengths included in the second combined image data, the second combined image data approaches the image quality of the inspection image, so it approaches the image quality of the correct answer image and the correct answer rate improves, but the image data As the amount increases, the processing load increases and the processing time increases. The image inspection system 100 can achieve both improved inspection accuracy and shortened inspection time with a suitable balance. In addition, the image inspection system 100 can efficiently detect color unevenness, etc. of the inspection object (second object) by using a wavelength with a high degree of contribution as the inspection wavelength (second wavelength). .

As described above, the image inspection system 100 of the present embodiment is an image inspection system that inspects an object using a multi-wavelength image, and the image inspection system 100 is an image inspection system that inspects an object using a multi-wavelength image. It includes an imaging unit (for example, camera 10) that receives light from the imaging unit to obtain a confirmation image GC, and a processing unit (for example, control device 30) that performs processing on the confirmation image GC acquired by the imaging unit. The processing section generates each image data gd of each wavelength constituting the multiple wavelengths included in the confirmation image GC, based on the confirmation image GC and the spectral characteristics of the spectroscopic section 11. The processing unit specifies a plurality of first wavelengths included in the multiple wavelengths, and generates first combined image data obtained by combining each image data gd of the specified first wavelengths and a predetermined image data corresponding to the first object. The correct answer rate of the first combined image data is calculated based on the correct answer image GA. When the correct answer rate is equal to or higher than the threshold th, the processing unit converts the plurality of second wavelengths used for inspection of the inspection image obtained by imaging the second object by the imaging unit into the plurality of specified first wavelengths. Decide on the wavelength.

As a result, the image inspection system 100 omit some of the wavelengths included in the confirmation image GC, generate the first combined image data, and compare it with the correct answer image GA, thereby increasing the correct answer rate. A plurality of first wavelengths forming the first combined image data satisfying the threshold value th or more can be determined. Therefore, when actually performing an image inspection of the second object, some of the wavelengths included in the inspection image are omitted to generate second combined image data and compared with the correct image GA. By doing so, the image inspection of the second object can be performed with the correct answer rate satisfying the threshold value th or more. In addition, since the second combined image data omits some of the wavelengths in the inspection image and the amount of image data is small, the processing load related to the comparison between the second combined image data and the inspection image is reduced. . Therefore, the image inspection system 100 can reduce the inspection time required for inspection while suppressing a decrease in inspection accuracy.

Further, the processing unit may specify areas a to d in which the correct answer rate is calculated in the confirmation image GC.

Thereby, the image inspection system 100 can specify which part of the confirmation image the correct answer rate or image inspection result is desired to obtain.

Further, the processing unit determines the correct answer rate for each wavelength based on the variation in the amount of received light for each wavelength obtained when the first object is imaged by the imaging unit according to each region of the first object. The degree of contribution may be calculated, and a plurality of first wavelengths may be designated based on the degree of contribution.

Thereby, the image inspection system 100 can determine the plurality of first wavelengths, taking into account the degree of contribution to the correct answer rate, that is, the ease of identifying each position of the first object.

Further, the processing unit may designate a plurality of first wavelengths by combining a plurality of wavelengths in descending order of contribution.

Thereby, the image inspection system 100 can determine, as the plurality of first wavelengths, a plurality of wavelengths that make it easy to identify each position of the first object.

Further, the processing unit may sequentially increase the number of wavelengths to be combined in descending order of contribution until the correct answer rate becomes equal to or greater than the threshold th.

Thereby, the image inspection system 100 can reduce the number of specified first wavelengths as much as possible while satisfying the desired correct answer rate in the image inspection. Therefore, the image inspection system 100 can reduce the processing load related to image inspection as much as possible.

Further, the processing section may specify a plurality of first wavelengths according to an input operation to the operation section (for example, the initial setting section 41 or the inspection condition setting section 42).

Thereby, the image inspection system 100 can manually specify the first wavelength desired by the user. For example, the image inspection system 100 can specify the first wavelength, taking into account the naturalness of the combined image data based on the specified first wavelength.

Further, the processing unit may specify the number of wavelengths of the specified first wavelength.

Thereby, the image inspection system 100 can generate combined image data by taking into account the number of designated first wavelengths.

Furthermore, the processing unit may specify a threshold th with which the correct answer rate is compared.

Thereby, the image inspection system 100 can adjust the inspection accuracy of the image inspection of the second object by adjusting the threshold th.

Further, the imaging section may receive light from the second object via the spectroscopic section 11 to obtain the inspection image. The processing section generates image data of each wavelength constituting the multi-wavelength based on the inspection image and the spectral characteristics of the spectroscopic section 11, and generates a second image data that is a combination of the image data of the plurality of determined second wavelengths. The second object may be inspected based on the combined image data and the correct image GA.

As a result, the image inspection system 100 uses the first wavelength specified using the confirmation image of the first object to inspect the inspection image obtained by capturing the second object. It can be a second wavelength. In other words, the image inspection system 100 can perform an image inspection using an inspection image with the same inspection accuracy and the same inspection time as an image inspection using a confirmation image.

The present disclosure is useful for an image inspection system, an image inspection method, and the like that can reduce the inspection time required for inspection while suppressing a decrease in inspection accuracy.

10 Camera 10A Camera unit 11 Spectroscopy section 12 Imaging section 13 Storage section 20 Illumination device 21 Light projection section 22 Storage section 30 Control device 30A Control unit 31 Control section 32 Calculation section 33 Judgment section 34 Storage section 40 Display operation device 41 Initial setting section 42 Inspection condition setting section 43 Image display section 44 Result display section 100 Image inspection system

Claims (10)

An image inspection system that inspects an object using multi-wavelength images,
an imaging unit that receives light from the first object and obtains a confirmation image via a spectroscopic unit that separates the light;
a processing unit that processes the confirmation image acquired by the imaging unit;
Equipped with
The processing unit includes:
generating each image data of each wavelength constituting the multiple wavelengths included in the confirmation image based on the confirmation image and the spectral characteristics of the spectroscopic section;
Specify a plurality of first wavelengths included in the multi-wavelength,
A correct answer rate of the first combined image data is determined based on the first combined image data obtained by combining each image data of the specified first wavelength and a predetermined correct answer image corresponding to the first target object. Calculate,
When the correct answer rate is equal to or higher than the threshold, the plurality of second wavelengths used for inspection of the inspection image obtained by imaging the second object by the imaging unit are set to the plurality of specified first wavelengths. decide,
Image inspection system. The processing unit specifies an area in the confirmation image for which the correct answer rate is to be calculated;
The image inspection system according to claim 1. The processing unit includes:
Based on the variation in the amount of light received for each wavelength obtained by imaging the first object by the imaging unit according to each region of the first object, the correct answer rate for each wavelength is calculated. Calculate the contribution,
specifying the plurality of first wavelengths based on the degree of contribution;
The image inspection system according to claim 1 or 2. The processing unit specifies the plurality of first wavelengths by combining the plurality of wavelengths in order of the contribution degree.
The image inspection system according to claim 3. The processing unit sequentially increases the number of wavelengths to be combined in descending order of contribution until the correct answer rate becomes equal to or greater than the threshold.
The image inspection system according to claim 4. The processing unit specifies the plurality of first wavelengths in response to an input operation to the operation unit.
The image inspection system according to claim 1 or 2. The processing unit specifies the number of wavelengths of the specified first wavelength,
The image inspection system according to any one of claims 1 to 6. The processing unit specifies the threshold value to be compared with the correct answer rate;
The image inspection system according to any one of claims 1 to 7. The imaging unit receives light from the second object via the spectroscopic unit to obtain the inspection image,
The processing unit includes:
generating each image data of each wavelength constituting multiple wavelengths based on the inspection image and the spectral characteristics of the spectroscopic section,
Inspecting the second object based on the correct image and second combined image data that is a combination of the determined image data of the plurality of second wavelengths;
The image inspection system according to any one of claims 1 to 8. An image inspection method for inspecting an object using multi-wavelength images, the method comprising:
an image acquisition step of receiving the light from the first object via a spectroscopic unit that spectrally separates the light and acquiring a confirmation image;
a processing step of performing processing on the obtained confirmation image;
has
The processing step includes:
generating image data of each wavelength constituting multiple wavelengths included in the confirmation image, based on the confirmation image and the spectral characteristics of the spectroscopic section;
specifying a plurality of first wavelengths included in the multi-wavelength;
A correct answer rate of the first combined image data is calculated based on the first combined image data obtained by combining each image data of the specified first wavelength and a predetermined correct answer image corresponding to the first object. a step of calculating;
If the correct answer rate is equal to or greater than a threshold, determining a plurality of second wavelengths to be used for inspecting an inspection image obtained by imaging a second object as the plurality of specified first wavelengths; ,
An image inspection method having

Images