Quality evaluation apparatus, quality evaluation system, and quality evaluation method

Abstract

To provide a quality evaluation apparatus, a system and a method for properly evaluating the quality of various workpieces, on the basis of color images including other elements, such as shadow or background, that are generated depending on the shapes of the workpieces.SOLUTION: In a quality evaluation system 1, a quality evaluation apparatus captures an image with visible light Lv, acquires a color image including a first region corresponding to a workpiece W, a second region corresponding to a foreign substance or an abnormal portion, and a third region corresponding of other elements, performs at least arithmetical operation on luminance values of pixels of first and second image data generated on the basis of the color image, to create determination image data, and detects a foreign substance 10a or an abnormal portion 10b on the basis of the determination image data. The first image data and the second image data are created on the basis of the color image, as a predetermined combination, so that a first luminance difference between the first region and the second region in the determination image data may be larger than a second luminance difference between the first region and the third region.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a quality evaluation apparatus, a quality evaluation system, and a quality evaluation method.

Background Art

[0002] For example, in the quality evaluation of a workpiece such as food shipped as a product, it is required to determine whether there are any foreign substances or abnormal parts on its appearance or cut surface. Conventionally, such quality evaluation has been performed manually by workers. However, when performing a full inspection of mass-produced products, etc., there are limitations. Therefore, automation has been promoted by analyzing an image obtained by imaging a workpiece using a camera.

[0003] Regarding such an automatic quality evaluation apparatus using image analysis technology, for example, there are Patent Documents 1 and 2. In Patent Document 1, food is handled as a workpiece, and a technique is disclosed in which, by performing binarization processing on an image obtained by imaging the food, it is determined that there is a foreign substance mixed in the food in a range exceeding a threshold value among adjacent pixels. Further, in Patent Document 2, although it is not a technique for evaluating foreign substances or abnormal parts, by performing binarization processing on a processed image obtained as the difference between an image using visible light and an image using infrared light for imaging a workpiece, an image processing technique for distinguishing and extracting the regions of the fat part and the bone part of meat is disclosed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] When performing quality evaluation of a workpiece by analyzing an image captured by a camera, the image may include, in addition to the normal part of the workpiece and foreign matter or abnormal parts that are the targets of quality evaluation, other elements such as shadows or backgrounds caused by the shape of the workpiece. In quality evaluation, it is important to distinguish such other elements and appropriately distinguish and evaluate the normal part of the workpiece and foreign matter or abnormal parts.

[0006] In each of the above patent documents, since other elements are not considered, there is a possibility that appropriate quality evaluation cannot be performed depending on the imaging environment of the image or the shape of the workpiece. Also, in Patent Document 1, quality evaluation based on a so-called monochrome image is performed by comparing the image with a threshold value for each pixel and performing binarization processing. Therefore, for example, the evaluation target is limited to a workpiece with a relatively large contrast in which white normal parts and black foreign matters are mixed. Further, in Patent Document 2, since two types of images using visible light and infrared light respectively are required, the time required for imaging, the computational load required for image processing, and the device cost increase.

[0007] At least one embodiment of the present disclosure has been made in view of the above circumstances, and an object thereof is to provide a quality evaluation device, a quality evaluation system, and a quality evaluation method capable of suitably evaluating the quality of various workpieces based on a color image including other elements such as shadows or backgrounds caused by the shape of the workpiece.

Means for Solving the Problems

[0008] In order to solve the above problems, a quality evaluation device according to at least one embodiment of the present disclosure is a quality evaluation device for detecting foreign matter or abnormal parts from a workpiece, a color image acquisition unit for acquiring a color image captured using visible light and including a first region corresponding to the workpiece, a second region corresponding to the foreign matter or the abnormal part, and a third region corresponding to other elements other than the workpiece; a first image data creation unit and a second image data creation unit for respectively creating first image data and second image data based on the color image; A determination image data creation unit for creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data; A detection unit for detecting the foreign matter or the abnormal part based on the determination image data; and The first image data and the second image data are created based on the color image as a predetermined combination such that a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region in the determination image data.

[0009] A quality evaluation system according to at least one embodiment of the present disclosure includes, in order to solve the above problems, a light source for irradiating the visible light; an imaging device for imaging the color image; a quality evaluation device according to at least one embodiment of the present disclosure; and

[0010] A quality evaluation method according to at least one embodiment of the present disclosure includes, in order to solve the above problems, a quality evaluation method for detecting a foreign matter or an abnormal part from a workpiece, acquiring a color image that is imaged using visible light and includes a first region corresponding to the workpiece, a second region corresponding to the foreign matter or the abnormal part, and a third region corresponding to other elements other than the workpiece; creating first image data and second image data respectively based on the color image; creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data; detecting the foreign matter or the abnormal part based on the determination image data; and The first image data and the second image data are respectively created based on the color image as a predetermined combination such that in the determination image data, a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region.

Advantages of the Invention

[0011] According to at least one embodiment of the present disclosure, it is possible to provide a quality evaluation apparatus, a quality evaluation system, and a quality evaluation method that can suitably evaluate various workpiece qualities based on a color image including other elements such as shadows and backgrounds caused by the shape of the workpiece.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Embodiments for Carrying Out the Invention

[0013] Hereinafter, some embodiments of the present invention will be described with reference to the accompanying drawings. However, the configurations described as embodiments or shown in the drawings are not intended to limit the scope of the present invention, but are merely illustrative examples.

[0014] First, referring to FIG. 1, a quality evaluation system 1 including a quality evaluation device 100 according to at least one embodiment of the present disclosure will be described. FIG. 1 is a schematic diagram of a quality evaluation system 1 according to one embodiment.

[0015] The quality evaluation system 1 is a system for evaluating the quality of a workpiece w. The workpiece w is conveyed in a predetermined conveyance direction a while being placed on the conveyor 2. The workpiece w is not limited, but for example, it is food.

[0016] The quality evaluation system 1 includes an imaging device 6, a light source 8, and a quality evaluation device 100. The imaging device 6 is configured to capture a color image Gc of the workpiece w using visible light Lv, and is, for example, a color camera. The visible light Lv used for imaging is irradiated from the light source 8 onto the workpiece w on the conveyor 2, and the reflected light Rv from the workpiece w is captured by the imaging device 6.

[0017] The light source 8 is installed at a position where it can irradiate the upper surface of the workpiece w conveyed on the conveyor 2 with visible light Lv. For example, the light source 8 is installed facing downward so as to irradiate the workpiece w with visible light Lv vertically above the conveyor 2 with respect to the conveyance direction a. Also, the imaging device 6 is installed facing the workpiece w conveyed by the conveyor 2 so as to easily receive the reflected light Rv from the workpiece w (more specifically, the imaging device 6 is installed vertically above the conveyor 2 with respect to the conveyance direction a and facing downward). When the visible light Lv necessary for imaging by the imaging device 6 can be sufficiently obtained as natural light, the light source 8 may be omitted in the quality evaluation system 1.

[0018] The quality evaluation device 100 is a device for performing quality evaluation of the work w using the color image captured by the imaging device 6. The quality evaluation by the quality evaluation device 100 includes the presence or absence of the foreign matter 10a or the abnormal part 10b in the work w, and further includes the detection of the foreign matter 10a or the abnormal part 10b when the foreign matter 10a or the abnormal part 10b is present. The foreign matter 10a is other substances (such as dust, dirt, etc.) mixed in the work w, and the abnormal part 10b is an area where a sound work w has mutated (such as a rotten part or a discolored part, etc.).

[0019] Such a quality evaluation device 100 is composed of, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and a computer-readable storage medium, etc. And a series of processes for realizing various functions are stored in a storage medium, etc. in the form of a program as an example. The CPU reads this program into the RAM, etc. and executes information processing and arithmetic processing to realize various functions. Incidentally, the program may be applied in a form pre-installed in the ROM or other storage media, a form provided in a state stored in a computer-readable storage medium, a form distributed via wired or wireless communication means, etc. A computer-readable storage medium is a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, etc.

[0020] FIG. 2 is a block configuration diagram of the quality evaluation device 100 in FIG. 1. The quality evaluation device 100 includes a color image acquisition unit 102, a first image data creation unit 104, a second image data creation unit 106, a determination image data creation unit 108, a detection unit 110, and a display unit 112.

[0021] The color image acquisition unit 102 is a configuration for acquiring the color image Gc captured by the imaging device 6. Each pixel of the color image Gc acquired by the color image acquisition unit 102 is represented by a luminance value in the RGB color space.

[0022] Here, FIG. 3 is a schematic diagram showing an example of the color image Gc acquired by the color image acquisition unit 102 in FIG. 2. The color image Gc includes a first region R1 corresponding to the sound part of the work w, a second region R2 corresponding to the foreign matter 10a or the abnormal part 10b, and a third region R3 corresponding to other elements. The other elements corresponding to the third region R3 are not limited. For example, there are the background of the work w reflected in the color image Gc (for example, the peripheral configuration such as the conveyor 2 shown in FIG. 1) and the shadow 12 generated on the work w due to the positional relationship with the light source 8.

[0023] In addition, in FIG. 1, the shadow 12 generated on the back side of the work w is exemplified as the shadow. However, since the position where the shadow 12 is generated varies depending on the shape of the work w, it is not limited to this. Further, when the work w does not have the foreign matter 10a or the abnormal part 10b, the color image Gc may not include the second region corresponding to the foreign matter 10a or the abnormal part 10b.

[0024] Returning to FIG. 2 again, the first image data creation unit 104 and the second image data creation unit 106 are configured to create the first image data G1 and the second image data G2. The first image data G1 and the second image data G2 are each created based on the color image Gc. In particular, as will be described later, when creating the determination image data Gj by performing arithmetic operations on the luminance values of each pixel of the first image data G1 and the second image data G2, the first luminance difference D1 between the first region R1 and the second region R2 in the determination image data Gj is made larger than the second luminance difference D2 between the first region R1 and the third region R3. They are each created based on the color image as a predetermined combination. That is, there are various types of image data that can be created based on the color image. However, in the determination image data Gj created when using the first image data and the second image data, a combination in which the first luminance difference D1 is larger than the second luminance difference D2 is predetermined. The first image data creation unit 104 and the second image data creation unit 106 each create image data corresponding to such a predetermined combination as the first image data and the second image data.

[0025] Also, the determination image data creation unit 108 is configured to create the determination image data Gj using the first image data G1 created by the first image data creation unit 104 and the second image data G2 created by the second image data creation unit 106. Specifically, the determination image data Gj is created by performing at least arithmetic operations (addition, subtraction, multiplication, division) on the first luminance value of the first image data G1 and the second luminance value of the second image data G2 for the luminance value of each pixel.

[0026] Here, referring to FIGS. 4 and 5, a method for creating the determination image data Gj based on such first image data G1 and second image data G2 will be specifically described. FIG. 4 is a diagram showing a plurality of image data based on the color image Gc obtained by the color image acquisition unit 102 of FIG. 2, and FIG. 5 is a diagram showing an example of creating the determination image data Gj based on the first image data G1 and the second image data G2 selected from FIG. 4.

[0027] So far, in this example, as the workpiece w, a potato having a cut surface is being handled. In particular, when the cut surface is placed upward on the conveyor 2, the color image Gc obtained by imaging the side including the cut surface by the imaging device 6 is handled.

[0028] As described above, in the color image Gc obtained by the color image acquisition unit 102, each pixel is represented as a luminance value in the RGB color space. The color image Gc in which each pixel is represented in the RGB color space can be separated into respective color component images (R component image, G component image, and B component image) as shown in the middle row of FIG. 4. Further, these RGB color spaces can be converted into other color spaces by a predetermined conversion operation. In the lower row of FIG. 4, respective color component images (H component image, S component image, and V component image) converted into the HSV color space are shown as an example of such other color spaces.

[0029] The first image data creation unit 104 and the second image data creation unit 106 can selectively create, as the first image data G1 and the second image data G2, image data of any component corresponding to a predetermined combination from each image data obtained based on such a color image Gc. Regarding which combination of image data is selected as the first image data G1 and the second image data G2, it may be determined in advance in consideration of the feature amounts (for example, shape, color, texture, area, center-of-gravity coordinates, width, height, aspect ratio, circularity, total length, rectangularity, radius, convexity, compactness, elliptical shape, etc.) of the workpiece w.

[0030] In one aspect, the first image data G1 may be any component in the RGB color space obtained based on the color image Gc, and the second image data G2 may be any component in another color space (for example, the HSV color space) obtained by converting the RGB color space. That is, in this case, the first image data G1 and the second image data G2 are obtained as image data corresponding to any component in different color spaces, respectively. Thereby, feature amounts different from RGB can be obtained, and feature amounts closer to human senses can be selected.

[0031] In another aspect, among the components of the RGB color space obtained based on the color image Gc, for the first image data G1, the image data of the color component that contributes the most to the color image Gc may be selected by principal component analysis (PCA: Principal Component Analysis). In this case, the principal component analysis may select the image data of the color component having the maximum luminance value among the components of the RGB color space. Generally, since the first region R1 corresponding to the normal part of the workpiece w has a wider range than the second region R2 corresponding to the foreign object 10a or the abnormal part 10b, by selecting the first image data G1 in this way, it is possible to obtain the determination image data Gj that can preferably evaluate the normal part of the workpiece w.

[0032] In yet another aspect, the second image data G2 may be a color component (for example, the S component of the HSV color space) in another color space obtained by converting the RGB color space obtained based on the color image Gc. Since the color components in such other color spaces generally correspond to the difference between the component having the maximum luminance and the component having the minimum luminance in the RGB color space, by selecting the second image data G2 in this way, it becomes possible to perform quality evaluation while appropriately considering factors that affect the contrast such as the shadow 12 included in the color image Gc.

[0033] In yet another aspect, both the first image data G1 and the second image data G2 may be selected as image data corresponding to any component of the RGB color space. Thereby, it is possible to select an image with high luminance required for binarization processing.

[0034] In FIG. 5, as one example of selection of such first image data G1 and second image data G2, the R component image in the RGB color space is selected as the first image data G1, and the B component image in the RGB color space is selected as the second image data G2. As an arithmetic operation, a state of creating determination image data Gj by subtracting them from each other is shown. In the determination image data Gj created based on the first image data G1 and the second image data G2 corresponding to a predetermined combination in this way, the first luminance difference D1 between the first region R1 corresponding to the normal part of the work w and the second region R2 corresponding to the abnormal part 10b is greater than the second luminance difference D2 between the first region R1 corresponding to the normal part of the work w and the third region R3 corresponding to the background (the wire mesh constituting the conveyor 2 on which the work w is placed) around the work w. Thus, determination image data Gj from which the influence of the background is eliminated (that is, the background is substantially removed) is obtained.

[0035] FIG. 6 is a diagram showing another creation example of the determination image data Gj corresponding to FIG. 5. In the example of FIG. 6, the S component image data in the HSV color space is selected as the first image data G1, and the G component image in the RGB color space is selected as the second image data G2. As an arithmetic operation, a state of creating determination image data Gj by subtracting them from each other is shown.

[0036] FIG. 7 is a diagram showing another creation example of the determination image data Gj corresponding to FIG. 5. In the example of FIG. 7, the R component image data, which is the principal component image data having the highest luminance value in the RGB color space, is selected as the first image data G1, and the B component image in the RGB color space is selected as the second image data G2. As an arithmetic operation, a state of creating determination image data Gj by subtracting them from each other is shown.

[0037] Returning to FIG. 2 again, the detection unit 110 is configured to detect the foreign object 10a or the abnormal part 10b based on the determination image data Gj created by the determination image data creation unit 108. In the detection unit 110, the presence or absence of such a foreign object 10a or abnormal part 10b, and when there is a foreign object 10a or abnormal part 10b, its shape, number, etc. can be detected. The detection of such a foreign object 10a or abnormal part 10b is performed, for example, by specifying the luminance value of each pixel constituting the determination image data Gj, and detecting the foreign object 10a or abnormal part 10b as a set of pixels having a luminance value equal to or greater than a preset reference value.

[0038] The display unit 112 is configured to output the detection result of the foreign object 10a or the abnormal part 10b by the detection unit 110. The output mode by the display unit 112 is not limited. For example, the detection result by the detection unit 110 is displayed on a display device such as a display for the operator of this device.

[0039] Here, FIG. 8 is an example of the display by the display unit 112 in FIG. 2. In this display example, for the color image Gc acquired by the imaging device 6, the second region R2 corresponding to the foreign object 10a or the abnormal part 10b detected by the detection unit 110 is displayed so as to be emphasized compared to the first region R1 and the third region R3. Thereby, the operator can suitably recognize the presence or absence and state of the foreign object 10a or the abnormal part 10b through the display of the display unit 112.

[0040] Subsequently, a quality evaluation method implemented by the quality evaluation device 100 having the above configuration will be described. FIG. 9 is a flowchart showing a quality evaluation method according to an embodiment.

[0041] First, the color image acquisition unit 102 acquires the color image Gc from the imaging device 6 (step S10). As described above, in this color image Gc, each pixel is specified by the luminance value of each component in the RGB color space.

[0042] Subsequently, when the first image data creation unit 104 and the second image data creation unit 106 create discrimination image data Gj based on the color image Gc acquired in step S10, they create the first image data G1 and the second image data G2 respectively as a predetermined combination such that the first luminance difference D1 is greater than the second luminance difference D2 (steps S12, S14). These first image data G1 and second image data G2 are input to the discrimination image data creation unit 108. The discrimination image data creation unit 108 creates discrimination image data Gj by performing arithmetic operations on the luminance values of the respective pixels of the first image data G1 and the second image data G2 (step S16).

[0043] Subsequently, the detection unit 110 detects the foreign object 10a or the abnormal part 10b based on the discrimination image data Gj created in step S16 (step S18). The detection result in step S18 is recognized by the operator of this apparatus by being displayed in a predetermined manner by the display unit 112 (step S20).

[0044] Subsequently, with reference to FIGS. 10 and 11, an evaluation example when handling a work w according to another example will be described. FIG. 10 shows another example of the color image Gc acquired by the color image acquisition unit 102 in FIG. 2, and the first image data G1 and the second image data G2 created based on the color image Gc, and FIG. 11 shows another example of the discrimination image data Gj created based on the first image data G1 and the second image data G2 in FIG. 10.

[0045] In this example, dried noodles are handled as the work w. The dried noodles are formed into a lump in a state where a plurality of noodles having an elongated shape are intertwined. In the color image Gc, in addition to the first region R1 corresponding to the healthy part of the work w and the second region R2 corresponding to the foreign object 10a or the abnormal part 10b (when there is a foreign object 10a or an abnormal part 10b), a third region R3 corresponding to the black shadow 12 reflected in the gaps between the noodles is included. Each pixel of such a color image Gc is represented by a luminance value in the RGB color space as described above.

[0046] Also, in this example, as the first image data G1, the R component image in the RGB color space among the color images Gc is selected, and as the second image data, the S component image in the HSV color space obtained by converting the RGB color space among the color images Gc is selected. In particular, the R color component image selected as the first image data G1 is the principal component (CPA) image having the maximum luminance value in the RGB color space.

[0047] FIG. 11 shows a state in which determination image data Gj is created based on the first image data G1 and the second image data G2 in FIG. 10. In this example, the determination image data Gj created by subtracting the second image data G2' obtained by extracting the region that becomes the shadow 12 from the second image data G2 from the first image data G1' obtained by extracting the dark portion from the first image data G1 is shown. In this determination image data Gj, the pixels corresponding to the foreign matter 10a or the abnormal portion 10b are shown in emphasis compared to the surroundings. Therefore, the detection unit 110 determines the presence or absence of the foreign matter 10a or the abnormal portion 10b in the workpiece w based on such determination image data Gj, and as a result, when there is the foreign matter 10a or the abnormal portion 10b, its position, number, shape, etc. can be detected.

[0048] In addition, within the scope not departing from the gist of the present disclosure, it is possible to appropriately replace the components in the above-described embodiments with well-known components, and the above-described embodiments may also be appropriately combined.

[0049] The content described in each of the above embodiments is understood as follows, for example.

[0050] (1) A quality evaluation apparatus according to one aspect is a quality evaluation apparatus for detecting foreign matter or an abnormal portion from a workpiece, a color image acquisition unit for acquiring a color image that is imaged using visible light and includes a first region corresponding to the workpiece, a second region corresponding to the foreign matter or the abnormal portion, and a third region corresponding to other elements other than the workpiece; A first image data creation unit and a second image data creation unit that respectively create first image data and second image data based on the color image; A determination image data creation unit for creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data; A detection unit for detecting the foreign matter or the abnormal part based on the determination image data; Comprising; The first image data and the second image data are created based on the color image as a predetermined combination such that a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region in the determination image data.

[0051] According to the aspect of (1) above, a color image including a first region to a third region is acquired by imaging using visible light. The first image data and the second image data created based on the acquired color image are used for creating determination image data by performing at least arithmetic operations on the luminance values of each pixel. In particular, the first image data and the second image data are created based on the color image as a predetermined combination such that a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region in the determination image data. By analyzing the determination image data created in this way, while appropriately distinguishing the third region included in the color image, based on the first region and the second region included in the color image, the presence or absence of foreign matter or abnormal parts in the workpiece can be discriminated, and the quality of the workpiece can be suitably evaluated.

[0052] (2) In another aspect, in the aspect of (1) above, The first luminance value at each pixel of the first image data and the second luminance value at each pixel of the second image data are luminance values corresponding to any component in the RGB color space or another color space converted from the RGB color space among the luminance values possessed by each pixel of the color image.

[0053] According to the aspect (2) above, the first image data and the second image are created such that the luminance values of the respective pixels of the first image data and the second image are luminance values corresponding to any one of the components in the RGB color space or another color space among the luminance values of the respective pixels of the color image. By appropriately selecting, according to the type of the work to be evaluated and other elements included in the color image, the luminance values corresponding to the components of each color space selected as the luminance values of the respective pixels of the first image data and the second image, the quality of the work can be suitably evaluated based on the color image. Note that the other color space is a color space other than the RGB space converted from the RGB space, and is, for example, any known color space including the HSV color space, the HSL color space, and the like.

[0054] (3) In another aspect, in the aspect (2) above, the first luminance value is a luminance value corresponding to any one of the components in the RGB color space among the luminance values of the respective pixels of the color image, and the second luminance value is a luminance value corresponding to any one of the components in the other color space among the luminance values of the respective pixels of the color image.

[0055] According to the aspect (3) above, the first image data is created such that the luminance value of the first image data is a luminance value corresponding to any one of the components in the RGB color space among the luminance values of the respective pixels of the color image, and the second image data is created such that the luminance value of the second image data is a luminance value corresponding to any one of the components in the other color space among the luminance values of the respective pixels of the color image. Thereby, a feature amount different from that of the RGB color space can be obtained, and a feature amount close to the human sense can be selected.

[0056] (4) In another aspect, in the aspect (3) above, the first luminance value is a luminance value corresponding to the component having the maximum luminance value of the color image in the RGB space.

[0057] According to the aspect (4) above, the first image data is created such that, as each pixel of the first image data, a luminance value corresponding to a component in the RGB color space having the maximum luminance value among the pixels of the color image is selected. Generally, since the first region corresponding to the normal part of the workpiece has a wider range than the second region corresponding to foreign matter or abnormal parts, by creating the first image data in this way, the normal part of the workpiece can be preferably evaluated.

[0058] (5) In another aspect, in the aspect (3) or (4) above, the second luminance value is a luminance value corresponding to a color component in the other color space among the luminance values of each pixel of the color image.

[0059] According to the aspect (5) above, the second image data is created such that, as each pixel of the second image data, a color component in the other color space (for example, the S component in the HSV color space) among the pixels of the color image is selected. Since the color component in the other color space generally corresponds to the difference between the component having the maximum luminance and the component having the minimum luminance in the RGB color space, by creating the second image data in this way, it becomes possible to perform a quality evaluation that appropriately considers elements that affect the contrast such as shadows included in the color image.

[0060] (6) In another aspect, in the aspect (2) above, the first luminance value and the second luminance value are luminance values corresponding to any one of the components in the RGB color space among the luminance values of each pixel of the color image.

[0061] According to the aspect (6) above, the first image data and the second image data are created such that the luminance values of the first image data and the second image data become luminance values corresponding to any one of the components in the RGB color space among the luminance values of each pixel of the color image. Thereby, it is possible to select an image with a high luminance required for binarization processing.

[0062] (7) In another aspect, in any one of the aspects (1) to (6) above, The workpiece is a food having a cut surface, The cut surface includes a mutated site that has occurred on the cut surface as the abnormal part.

[0063] According to the aspect (7) above, by detecting a mutated site on the cut surface as an abnormal part based on a color image obtained by imaging a food having a cut surface, quality evaluation can be suitably performed.

[0064] (8) In another aspect, in any one of the aspects (1) to (7) above, The other element is a shadow caused by the shape of the workpiece or the background of the workpiece.

[0065] According to the aspect (8) above, based on a color image including other elements such as a shadow or a background caused by the shape of the workpiece, the quality of various workpieces can be suitably evaluated.

[0066] (9) A quality evaluation system according to one aspect includes, A light source for irradiating visible light, An imaging device for imaging the color image, The quality evaluation device according to any one of the aspects (1) to (8) above, and is provided with.

[0067] According to the aspect (9) above, based on a color image captured using visible light, the quality of various workpieces can be suitably evaluated.

[0068] (10) A quality evaluation method according to one aspect is A quality evaluation method for detecting foreign matter or abnormal parts from a workpiece, A step of obtaining a color image that is imaged using visible light and includes a first region corresponding to the workpiece, a second region corresponding to the foreign matter or the abnormal part, and a third region corresponding to other elements other than the workpiece, A step of respectively creating first image data and second image data based on the color image, A step of creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data respectively; A step of detecting the foreign matter or the abnormal part based on the determination image data; comprising; The first image data and the second image data are respectively created based on the color image as a predetermined combination such that in the determination image data, a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region.

[0069] According to the aspect of (10) above, a color image including a first region to a third region is acquired by imaging using visible light. The first image data and the second image data created based on the acquired color image are used for creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of each. In particular, the first image data and the second image data are created based on the color image as a predetermined combination such that in the determination image data, a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region. By analyzing the determination image data created in this way, while appropriately distinguishing the third region included in the color image, based on the first region and the second region included in the color image, the presence or absence of foreign matter or abnormal parts in the work can be discriminated, and the quality of the work can be preferably evaluated.

Explanation of Signs

[0070] 1 Quality evaluation system 2 Conveyor 6 Imaging device 8 Light source 10a Foreign matter 10b Abnormal part 100 Quality evaluation device 102 Color image acquisition unit 104 First image data creation unit 106 Second image data creation unit 108 Determination image data creation unit 110 Detection unit 112 Display unit w Workpiece Lv Visible light Gc Color image Rv Reflected light

Claims

1. A quality evaluation apparatus for detecting foreign matter or abnormal parts from a workpiece, a color image acquisition unit for acquiring a color image that is captured using visible light and includes a first region corresponding to the workpiece, a second region corresponding to the foreign matter or the abnormal part, and a third region corresponding to other elements other than the workpiece; a first image data creation unit and a second image data creation unit for respectively creating first image data and second image data based on the color image; a determination image data creation unit for creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data; a detection unit for detecting the foreign matter or the abnormal part based on the determination image data; comprising: The first image data and the second image data are respectively created based on the color image as a predetermined combination such that a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region in the determination image data. Quality evaluation apparatus.

2. The first luminance value of each pixel of the first image data and the second luminance value of each pixel of the second image data are luminance values corresponding to any one of the components in the RGB color space or another color space converted from the RGB color space among the luminance values of each pixel of the color image. The quality evaluation apparatus according to claim 1.

3. The first luminance value is a luminance value corresponding to any one of the components in the RGB color space among the luminance values of each pixel of the color image, The second luminance value is a luminance value corresponding to any one of the components in the other color space among the luminance values of each pixel of the color image. The quality evaluation apparatus according to claim 2.

4. The first luminance value is a luminance value corresponding to the component having the maximum luminance value in the RGB space in the color image. The quality evaluation apparatus according to claim 3.

5. The second luminance value is a luminance value corresponding to the color component among the other color spaces among the luminance values of each pixel of the color image. The quality evaluation apparatus according to claim 3.

6. The first luminance value and the second luminance value are luminance values corresponding to any one of the components in the RGB color space among the luminance values of each pixel of the color image. The quality evaluation apparatus according to claim 2.

7. The workpiece is a food having a cut surface, The cut surface includes a mutated site generated on the cut surface as the abnormal part, and the quality evaluation apparatus according to claim 1 or 2.

8. The other element is a shadow generated by the shape of the workpiece or the background of the workpiece, and the quality evaluation apparatus according to claim 1 or 2.

9. A light source for irradiating the visible light, An imaging device for imaging the color image, The quality evaluation apparatus according to any one of claims 1 to 8, A quality evaluation system comprising the same.

10. A quality evaluation method for detecting a foreign object or an abnormal part from a workpiece, A step of acquiring a color image including a first region corresponding to the workpiece, a second region corresponding to the foreign object or the abnormal part, and a third region corresponding to other elements other than the workpiece, which is imaged using visible light; A step of creating first image data and second image data respectively based on the color image; A step of creating determination image data by performing at least arithmetic operations on the luminance values of each pixel of the first image data and the second image data; A step of detecting the foreign object or the abnormal part based on the determination image data; Comprising, The first image data and the second image data are created based on the color image as a predetermined combination such that a first luminance difference between the first region and the second region is greater than a second luminance difference between the first region and the third region in the determination image data. Quality evaluation method.

Images