Color unevenness evaluation device, color unevenness evaluation system, color unevenness evaluation method and program

Abstract

To enable accurate evaluation of color unevenness.SOLUTION: A color unevenness evaluation device comprises: an imaging portion that captures an image of an object; a projected member that is projected on a surface of the object; and an output portion that outputs a color unevenness evaluation value of the object acquired on the basis of the image captured by the imaging portion. The captured image includes an image of the projected member that is projected on the surface of the object.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a color unevenness evaluation device, a color unevenness evaluation system, a color unevenness evaluation method, and a program. [Background technology]

[0002] In the quality evaluation of a product, surface defects such as scratches, bumps, or orange peel, or color defects on a painted surface called color unevenness, etc., are often evaluated. A configuration is known in which such defect evaluation is performed based on images captured by an imaging unit such as a camera.

[0003] Also disclosed is a configuration that includes an imaging unit having an area camera, multiple planar lighting devices arranged at a certain angle with the axial direction of the area camera, and a diffusion plate arranged around the lens of the area camera for diffusing the illumination light from the planar lighting devices, and an articulated robot having a function of controlling the position and angle of the imaging unit relative to an object to be imaged, and that detects defects in the object from an image obtained by irradiating the object with illumination light and imaging it (see, for example, Patent Document 1). Summary of the Invention [Problem to be solved by the invention]

[0004] However, with the configuration of Patent Document 1, there are cases where color unevenness cannot be evaluated accurately.

[0005] An object of the present invention is to enable accurate evaluation of color unevenness. [Means for solving the problem]

[0006] The color unevenness evaluation device according to one aspect of the present invention is Imaging Image acquisition an imaging unit that projects a projection image onto a surface of the object; before Output the color unevenness evaluation value of the object process and The processing unit includes an image of the projection target projected onto a surface of the object. Based on the captured image, an ideal image of the object without color unevenness is obtained. and generating a difference image between the captured image and the ideal image based on the sum of absolute values ​​of pixel values ​​included in the difference image.The color unevenness evaluation value is output. Effect of the Invention

[0007] According to the present invention, color unevenness can be accurately evaluated. [Brief description of the drawings]

[0008] [Figure 1] 1 is a diagram showing an example of the overall configuration of a color unevenness evaluation device according to a first embodiment. [Diagram 2] 4 is a block diagram showing an example of a functional configuration of a processing unit; [Diagram 3] 10 is a flowchart illustrating an example of processing by a processing unit according to the embodiment. [Figure 4] 11A and 11B are diagrams illustrating an example of an extraction process of an evaluation image. [Diagram 5] 5A and 5B are diagrams showing an example of filter processing, where FIG. 5A is an image before processing and FIG. 5B is an image after processing. [Figure 6] 6A and 6B are diagrams showing an example of a process for generating an ideal image, where FIG. 6A is the original image and FIG. 6B is the ideal image. [Figure 7] 7A and 7B are diagrams showing an example of image difference processing, where FIG. 7A is the original image, FIG. 7B is the ideal image, and FIG. 7C is the difference image. [Figure 8] FIG. 11 is a diagram showing a first example of an accuracy verification result. [Figure 9] FIG. 13 is a diagram showing a second example of the accuracy verification result. [Figure 10] FIG. 13 is a diagram showing an example of the configuration of a color unevenness evaluation device according to a first modified example of the first embodiment. [Figure 11] FIG. 13 is a diagram showing an example of the configuration of a color unevenness evaluation device according to a second modified example of the first embodiment. [Figure 12] 12A and 12B are diagrams showing a modified example of processing by an extraction unit, in which FIG. 12A is a diagram before processing and FIG. 12B is a diagram after processing. [Figure 13] FIG. 13 is a diagram illustrating an example of the configuration of a color unevenness evaluation device according to a second embodiment. [Figure 14] FIG. 13 is a diagram showing a first example of the configuration of a color unevenness evaluation device according to a third embodiment. [Figure 15]FIG. 13 is a diagram showing a second example of the configuration of a color unevenness evaluation device according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In each drawing, the same components are denoted by the same reference numerals, and duplicated explanations will be omitted as appropriate.

[0010] The following embodiments are illustrative of a color unevenness evaluation device for embodying the technical concept of the present invention, and the present invention is not limited to the following embodiments. The dimensions, materials, shapes, relative positions, etc. of the components described below are intended to be illustrative, and not to limit the scope of the present invention, unless otherwise specified. Furthermore, the sizes and positional relationships of the components shown in the drawings may be exaggerated to clarify the explanation.

[0011] The color unevenness evaluation device according to the embodiment includes an imaging section that captures an image of an object, and an evaluation section that evaluates color unevenness of the object based on the captured image.

[0012] Here, color unevenness refers to a defect in which there is an area on a colored surface that is different in brightness or color from the original color. There are various types of color unevenness, such as areas of different colors that are banded or streaked, or areas where the colors appear vaguely different, such as haze. These types of color unevenness are also called color unevenness, color skewing, mottling, or mottle.

[0013] The object to be evaluated for color unevenness is not particularly limited, but may be, for example, a three-dimensional object such as an automobile. The color unevenness evaluation device according to the embodiment can evaluate color unevenness in the paint color applied to the body of the automobile.

[0014] For example, when evaluating color unevenness, images of objects around the target object may be reflected on the target object's surface. This reflection is more noticeable for targets with high gloss. This reflection may cause brightness unevenness within the image area of ​​the target object included in the captured image, making it difficult to accurately evaluate color unevenness.

[0015] Furthermore, uneven illuminance may occur in a captured image depending on the surface shape of a three-dimensional object, making it difficult to accurately evaluate color unevenness.

[0016] The color unevenness evaluation device according to the embodiment has a projection member that is projected onto the surface of the object, and the captured image includes an image of the projection member projected onto the surface of the object. By evaluating color unevenness based on this captured image, it is possible to suppress the image of objects around the object from being transferred onto the surface of the object, and the influence of illuminance unevenness, etc., and to accurately evaluate color unevenness.

[0017] [First embodiment] First, a color unevenness evaluation device 100 according to a first embodiment will be described.

[0018] <Configuration Example of Color Unevenness Evaluation Device 100> (Overall configuration example) FIG. 1 is a diagram showing an example of the overall configuration of a color unevenness evaluation device 100. As shown in FIG.

[0019] As shown in FIG. 1, the color unevenness evaluation device 100 includes an illumination section 1, an imaging section 2, a projection target member 3, and a processing section 4.

[0020] The color unevenness evaluation device 100 captures an image of the object S with the imaging unit 2 while the illumination unit 1 is illuminating the object S with light. When capturing an image, the positional relationship between the object S, the illumination unit 1, the imaging unit 2, and the projection target member 3 is determined in advance so that the image of the projection target member 3 is reflected on the surface of the object S. The color unevenness evaluation device 100 evaluates the color unevenness of the object S with the processing unit 4 based on the captured image captured by the imaging unit 2.

[0021] The illumination unit 1 includes a light source and illuminates the target object S with light emitted by the light source. In this embodiment, a TLWA325X-22WD-4 manufactured by ITEC Systems Co., Ltd. that has an LED (Light Emitting Diode) as the light source and emits white diffused light is used as an example.

[0022] The imaging unit 2 has an area camera and a lens, and captures an image of the object S approximately formed by the lens with the area camera. In this embodiment, the STC-MCS241U3V manufactured by Omron Sentec, which is a color area camera with a Bayer array, is used as an example of the area camera. The FL-CC1214-2M manufactured by Ricoh Industrial Solutions Inc., which has a focal length of 12 mm, is used as an example of the lens.

[0023] The projection target member 3 is a member that is projected onto the surface of the object S. An image of the projection target member 3 is projected onto the surface of the object S. When viewed from the imaging unit 2, this image of the projection target member 3 is a virtual image of the projection target member 3 reflected on the surface of the object S.

[0024] In this embodiment, a white film #188-E20 manufactured by Toray Industries, Inc. is used as an example of the projection target 3. It is desirable for the projection target 3 to be of a generally uniform color overall. However, it is not limited to white, and may be of a color other than white as long as it is of a generally uniform color.

[0025] 1, the illumination unit 1 is disposed with respect to the object S so that the illumination angle θ1, which is the angle between the central axis 11 of the illumination unit 1 and the central axis S1 of the object S, is approximately 35 degrees. The central axis 11 of the illumination unit 1 corresponds to the central axis of the LED that is approximately perpendicular to the light-emitting surface of the LED.

[0026] The imaging unit 2 is disposed with respect to the object S so that the imaging angle θ2, which is the angle between the central axis 21 of the imaging unit 2 and the central axis S1 of the object S, is approximately 45 degrees. The central axis 21 of the imaging unit 2 corresponds to the central axis of the imaging surface that is approximately perpendicular to the imaging surface of the area camera.

[0027] The projection target 3 is disposed with respect to the object S so that the projection angle θ3, which is the angle between the central axis 31 that is approximately perpendicular to the flat surface of the projection target 3 and the central axis S1 of the object S, is approximately 45 degrees. The projection target 3 is disposed on the opposite side of the central axis S1 of the object S to the illumination unit 1.

[0028] The working distance, which corresponds to the distance between the object S and the imaging unit 2, is approximately 380 [mm]. The resolution of the image captured by the imaging unit 2 is approximately 200 [μm / pixel]. The resolution of the captured image means the length of the image corresponding to the length of one side of each pixel constituting the captured image.

[0029] The processing unit 4 evaluates color unevenness of the object S based on the image captured by the imaging unit 2. The processing unit 4 is, for example, an arithmetic processing device implemented by a computer.

[0030] The color unevenness evaluation device 100 acquires a captured image as 12-bit RAW data. The color unevenness evaluation device 100 also acquires image data of the captured image by performing demosaicing using bilinear interpolation for each of the colors red (R), green (G), and blue (B) and calculating the luminance value of each color.

[0031] The arrangement of the illumination unit 1, imaging unit 2, and projection target member 3 is not particularly limited and can be appropriately selected, but in order for the projection target member 3 to be projected onto the surface of the target object S, it is preferable that the projection target member 3 is installed in the regular reflection direction of the imaging unit 2. In other words, it is preferable that the projection target member 3 is installed on the opposite side of the imaging unit 2 across the central axis S1 of the target object S, and that the imaging angle θ2 and the projection angle θ3 are approximately equal in value. In addition, it is preferable that the illumination unit 1 is installed in a position that is not included in the image of the target object S captured by the imaging unit 2.

[0032] (Action of the projection target 3) Here, the effect of the image captured by the imaging section 2 including the image of the projection target 3 projected on the surface of the target object S will be described.

[0033] The first effect is to prevent the images of objects such as walls and interior lighting fixtures around the object S from being reflected on the surface of the object S. If the object S is an automobile body, the painted surface of the automobile body is clear-coated and highly glossy. Therefore, the painted surface may act as a reflecting mirror, causing the images of surrounding objects to be reflected on the painted surface.

[0034] If an image of the car body is captured in this state, the image reflected on the painted surface makes it difficult to accurately extract and evaluate only the color unevenness on the body.

[0035] By providing the projection member 3, images of objects such as walls and indoor lighting fixtures that exist around the target object S can be blocked, preventing these images from being projected onto the surface of the target object S. Note that, although the case where the target object S is the body of an automobile has been described as an example, the same applies to target objects S other than the body of an automobile.

[0036] The second effect is to suppress uneven illuminance of the light that the illumination unit 1 illuminates the object S. If the object S is a three-dimensional object with a curved surface, the illumination light may be reflected on the surface of the object S to create a shadow, resulting in uneven illuminance on the surface of the object S. If the object S is imaged in this state, it becomes difficult to accurately extract and evaluate only the color unevenness of the object S due to the influence of uneven illuminance.

[0037] The effects of such uneven illuminance can be suppressed by projecting the image of the projection target 3, which has a substantially uniform color, onto the surface of the object S and using a captured image of the object S on whose surface the image of the projection target 3 is projected.

[0038] In order to suppress illuminance unevenness, instead of providing a projection target member 3, it is also possible to configure the illumination light itself, which has a substantially uniform color, to be projected onto the surface of the object S. However, in this configuration, the specular reflection light component of the illumination light by the surface of the object S becomes large, so that when the surface of the object S is a dark color such as black, the difference between the actual color of the surface of the object S and the color of the part with color unevenness becomes small in the captured image. As a result, it becomes difficult to accurately extract and evaluate only the color unevenness.

[0039] In this embodiment, by providing an illumination unit 1 that illuminates the object S with light and a projection target member 3 separate from the illumination unit 1, it becomes possible to accurately extract and evaluate color unevenness even when the object S has a dark color.

[0040] (Example of functional configuration of processing unit 4) Fig. 2 is a block diagram showing an example of a functional configuration of the processing unit 4 included in the color unevenness evaluation device 100. As shown in Fig. 4, the processing unit 4 has an extraction unit 41, an ideal image generation unit 42, a difference image acquisition unit 43, an evaluation unit 44, and an output unit 45.

[0041] These functions are realized by electric circuits, and some of these functions can also be realized by software (CPU: Central Processing Unit). These functions may also be realized by multiple circuits or multiple pieces of software. In order to perform image processing, it is preferable that the processing unit 4 has a frame memory such as a RAM (Random Access Memory) and is capable of temporarily storing image data of captured images, etc.

[0042] The processing unit 4 generates an ideal image free of color unevenness from the image captured by the imaging unit 2 using an ideal image generating unit 42, and extracts image components indicative of color unevenness from a difference image between the captured image and the ideal image using a difference image acquiring unit 43. The processing unit 4 acquires a color unevenness evaluation value by calculation using an evaluating unit 44 based on the size of the extracted image components. The processing unit 4 is able to acquire the color unevenness evaluation value only from the captured image of the object S, without capturing an image of a reference sample that serves as a standard free of color unevenness.

[0043] The extraction section 41 extracts, from the images captured by the imaging section 2, an evaluation image in which the image of the projection target member 3 is projected.

[0044] The ideal image generating section 42 generates an ideal image of the object S without color unevenness by calculation based on the image captured by the imaging section 2.

[0045] The difference image acquisition section 43 acquires a difference image between the image captured by the imaging section 2 and the ideal image generated by the ideal image generation section 42 through calculation.

[0046] The evaluation unit 44 obtains a color unevenness evaluation value of the object S by calculation based on the image captured by the imaging unit 2. For example, the evaluation unit 44 obtains a color unevenness evaluation value of the object S based on the sum of absolute values ​​of pixel values ​​included in the differential image captured by the differential image acquisition unit 43. The evaluation unit 44 can also correct the color unevenness evaluation value according to the chromaticity value of the object S.

[0047] The evaluation unit 44 outputs the acquired color unevenness evaluation value to an external device via the output unit 45. Examples of the external device include a display device such as a PC (Personal Computer) or a display, a storage device such as a HDD (Hard Disk Drive), or a communication device such as a communication I / F (Interface), which are provided outside the color unevenness evaluation device 100.

[0048] <Operation Example of Color Unevenness Evaluation Device 100> Next, the operation of the color unevenness evaluating apparatus 100 will be described.

[0049] (Example of how to capture an image) First, for calibration, the color unevenness evaluation device 100 captures an image of a white reference plate placed at the position of the target object S with the imaging unit 2 while the illumination unit 1 is turned on. The exposure time of the imaging unit 2 when capturing an image of the white reference plate is Tw [sec]. The color unevenness evaluation device 100 calculates the average luminance values ​​Rw, Gw, and Bw of the white reference plate from the captured image of the white reference plate.

[0050] Next, an object S is placed in place of the white reference plate, and an image of the object S is captured. The exposure time for capturing an image of the object S is Ts [sec], and the luminance values ​​of the R, G, and B colors of the captured image of the object S are Rs(i,j), Gs(i,j), and B(i,j), respectively. Note that i and j represent coordinates indicating a position in the captured image.

[0051] Next, the color unevenness evaluation device 100 corrects the captured image of the object S using the luminance value of the white reference plate. For example, the color unevenness evaluation device 100 performs the correction by calculating R(i,j), G(i,j), and B(i,j), which are the luminance values ​​for each color of each pixel in the corrected captured image of the object S, using the following formula (1).

number

[0052] The process for the above calibration may be performed by any of the functional components shown in Fig. 2. Furthermore, the color unevenness evaluating device 100 may have a separate functional component for the calibration process in the processing unit 4.

[0053] (Example of processing method by processing unit 4) Fig. 3 is a flowchart showing an example of processing by the processing unit 4. Fig. 3 shows processing triggered by the timing when an image captured by the imaging unit 2 is input to the processing unit 4.

[0054] First, in step S31, the extraction unit 41 extracts an evaluation image in which the image of the projection target 3 is reflected, from among the images captured by the imaging unit 2.

[0055] Next, in step S32, the ideal image generating unit 42 generates an ideal image of the object S without color unevenness based on the image captured by the imaging unit 2 by calculation.

[0056] Next, in step S33, the difference image acquisition section 43 acquires a difference image between the image captured by the imaging section 2 and the ideal image generated by the ideal image generation section 42 by calculation.

[0057] Next, in step S34, the evaluation unit 44 obtains a color unevenness evaluation value of the object S by calculation based on the difference image.

[0058] Next, in step S35, the output unit 45 outputs the color unevenness evaluation value acquired by the evaluation unit 44 to an external device.

[0059] In this way, the processing unit 4 can obtain a color unevenness evaluation value and output the obtained color unevenness evaluation value to an external device.

[0060] (Detailed example of processing method) Next, the processing method performed by the processing unit 4 will be described in detail.

[0061] Fig. 4 is a diagram illustrating an example of the extraction process of the evaluation image by the extraction unit 41. The captured image 7 shown in Fig. 4 is an image captured by the imaging unit 2. The captured image 7 includes an image 70 of the projection target member 3 projected on the surface of the target object S.

[0062] The extraction unit 41 extracts, by trimming, an evaluation image 71 for evaluating color unevenness from an image region corresponding to an image 70 of the projection target member 3 included in the captured image 7. The size of the evaluation image 71 may be determined in advance, or an appropriate range may be automatically extracted from within the captured image 7.

[0063] Next, the processing unit 4 performs a filter process on the evaluation image 71 extracted by the extraction unit 41. This filter process may be performed by any of the functional configuration units shown in Fig. 2, but here it will be described as being performed by the extraction unit 41.

[0064] Fig. 5 is a diagram for explaining an example of filter processing on an evaluation image 71. Fig. 5(a) is a diagram showing the evaluation image 71 before filter processing, and Fig. 5(b) is a diagram showing an evaluation image 71' after filter processing.

[0065] 5(a) and 5(b) are, for example, square images with 70 pixels on each side, which corresponds to a length of approximately 14 mm.

[0066] For example, if the object S is the body of an automobile and the body is coated with a metallic or pearlescent paint, the paint contains a lustrous material, and fine bright spots 72 corresponding to particle images will appear in the evaluation image 71, as shown in Figure 5(a).

[0067] Since the luminance fluctuation of the bright spots 72 caused by particles is not color unevenness, it is preferable to remove or make them less noticeable when evaluating color unevenness. The extraction unit 41 removes or makes the bright spots 72 less noticeable by filtering using, for example, a median filter. FIG. 5(b) shows the result of filtering using a median filter, which is a square filter with 11 pixels on each side. As shown in FIG. 5(b), the bright spots 72 have been removed.

[0068] Any filtering method may be used as long as it is capable of suppressing particle images such as the bright spots 72 in the evaluation image 71.

[0069] Next, the processing unit 4 performs a conversion process to chromaticity values ​​on the evaluation image 71' after the filtering process has been performed by the extraction unit 41. This conversion process is a process for converting RGB values, which are a device-dependent format, into L*a*b values, which are a uniform color space. This conversion process may be performed by any of the functional configuration units shown in Fig. 2, but in this example, it is described as being performed by the extraction unit 41.

[0070] The extraction unit 41 converts an image expressed in RGB values ​​into an image expressed in XYZ values ​​using the conversion formula shown in the following equation (2), and then converts the image expressed in XYZ values ​​into an image expressed in L*a*b values ​​using the conversion formula shown in the following equation (3).

number

number

[0071] In addition, Xn, Yn, and Zn in formula (3) represent the tristimulus values ​​X, Y, and Z of a perfect diffuse reflection surface. In this embodiment, the values ​​used are Xn=94.81, Yn=100.00, and Zn=107.33 under a D65 light source and a 10-degree viewing angle.

[0072] Next, the processing unit 4 causes the ideal image generating unit 42 to generate an ideal image, which is an image without color unevenness, from the image expressed by L* after conversion processing into chromaticity values.

[0073] Fig. 6 is a diagram for explaining an example of the generation process of an ideal image. Fig. 6(a) shows the original image, and Fig. 6(b) shows the ideal image. The original image is an image expressed by L* in the image after conversion process to chromaticity values.

[0074] In Fig. 6(a), color unevenness M is an image region that includes colors that are different from the uniform or continuous color changes in original image 73. Based on original image 73, ideal image generating unit 42 generates ideal image 74, which is an image in which colors change smoothly and does not include color unevenness M, as shown in Fig. 6(b).

[0075] In this embodiment, the ideal image generating unit 42 performs fitting with a polynomial function to generate an ideal image 74 that is free of color unevenness M. Specifically, the ideal image generating unit 42 performs fitting with a cubic function for each row in the horizontal direction of the original image 73, and then performs fitting in the same manner for each column in the vertical direction of the original image 73 to generate the ideal image 74.

[0076] Next, the processing unit 4 performs image difference processing from the original image 73 to the ideal image 74 using the difference image acquisition unit 43 to acquire a difference image.

[0077] Fig. 7 is a diagram for explaining an example of image difference processing. Fig. 7(a) shows an original image 73, Fig. 7(b) shows an ideal image 74, and Fig. 7(c) shows a difference image 75. The difference image acquisition unit 43 acquires a difference image 75 containing only the components of color unevenness M by the image difference between the original image 73 and the ideal image 74, and extracts only the color unevenness components.

[0078] Next, the processing unit 4 acquires a color unevenness evaluation value by calculation using the evaluation unit 44. The following formula (4) is a formula for calculating the color unevenness evaluation value.

number

[0079] The larger the brightness difference of the color unevenness and the larger the area of ​​the color unevenness, the larger the color unevenness evaluation value. The color unevenness evaluation value E1 obtained using formula (4) corresponds to an example of the color unevenness evaluation value obtained based on the sum of the absolute values ​​of the pixel values ​​included in the difference image.

[0080] Furthermore, as a result of careful investigation by the inventors, it was found that the correlation with human subjective evaluation can be increased by using the average L* of the object S as a correction variable for the total value of color unevenness obtained from the difference image 75.

[0081] The formula (5) used by the evaluation unit 44 to calculate the corrected color unevenness evaluation value E2 after correcting the influence of the color of the object S is shown in Equation (5). In Equation (5), p is a parameter that is determined in advance based on the subjective evaluation score. The corrected color unevenness evaluation value E2 corresponds to an example of a color unevenness evaluation value corrected according to the chromaticity value of the object S.

number

[0082] <Example of verification results of evaluation accuracy using color unevenness evaluation device 100> Next, the results of verification of the evaluation system using the color unevenness evaluation device 100 will be described.

[0083] (Verification method) The object S for verification was prepared by painting the surface of a resin steering wheel with a commercially available automotive repair paint. Three types of object S were prepared, each with three metallic colors: silver, red, and blue. The degree of color unevenness was intentionally varied by changing the number of coats applied and the drying time for each color. The number of objects S was four for silver, four for red, and five for blue.

[0084] Using the above-mentioned object S, the evaluation results by the color unevenness evaluation device 100 were compared with the subjective evaluation results, which are sensory evaluations by humans, to verify the correlation between the color unevenness evaluation values ​​by the color unevenness evaluation device 100 and the subjective evaluation scores.

[0085] In the subjective evaluation, a state in which there was absolutely no color unevenness was assigned an evaluation level of 0, and the object S with the worst degree of color unevenness in the inventor's subjective opinion was assigned an evaluation level of 10 as the reference object. Each of six evaluators other than the inventor was asked to observe object S and evaluate the degree of color unevenness in increments of 0.5 between evaluation levels 0 and 11.

[0086] The evaluators observed the object S under a D65 light source, using a Spectra Light QC manufactured by X-rite as the light source. The evaluation results from each of the six evaluators were used to calculate the average of the evaluation levels given to each object S, which was used as the subjective evaluation score for each object S.

[0087] Fig. 8 is a diagram showing a first example of accuracy verification results. The horizontal axis of the graph shown in Fig. 8 indicates the corrected color unevenness evaluation value E2 obtained by formula (5), and the vertical axis indicates the subjective evaluation score. The color unevenness of each color, silver, red, and blue, is the legend. The error bar on the vertical axis indicates the standard deviation of the subjective evaluation score. Also, the parameter p in formula (5) is 0.44, and is calculated based on the subjective evaluation score.

[0088] The results shown in Fig. 8 show that the contribution rate is 0.9 or more, and that there is a high correlation between the corrected color unevenness evaluation value E2 obtained by the color unevenness evaluation device 100 and the evaluation result by subjective evaluation. In addition, the results shown in Fig. 8 also show that color unevenness is mainly caused by brightness fluctuations.

[0089] Here, Fig. 9 is a diagram showing a second example of the accuracy verification result. Fig. 9 can be interpreted in the same way as Fig. 8. Fig. 9 shows the result when the average L*, which is the correction variable, is not used, that is, when the color unevenness evaluation value E1 calculated using formula (4) is used as the color unevenness evaluation value. It was found that when the color unevenness evaluation value E1 is used, differences in the correlation occur depending on the color of the target object S.

[0090] <Effects of the color unevenness evaluation device 100> Next, the effects of the color unevenness evaluating apparatus 100 will be described.

[0091] One of the quality control items for products is the appearance of the surface. In general, appearance is evaluated by looking at surface defects such as scratches, bumps, or orange peel, or color defects on the painted surface called color unevenness. These appearance evaluations are often performed by human visual inspection.

[0092] In recent years, in order to reduce the subjectivity of evaluation that requires an experienced evaluator, the automation of product appearance evaluation has been promoted for the purpose of improving the manufacturing process by quantifying the evaluation results, or reducing manufacturing costs by reducing the number of personnel, etc. As part of such automation, a configuration is known in which the appearance is photographed using an imaging unit such as a camera, and defects are evaluated based on the photographed image.

[0093] However, in a conventional configuration using captured images, it can be difficult to evaluate color unevenness in some cases. For example, when an image of an object existing around the target object is reflected in the captured image, or when the captured image has illuminance unevenness due to the shape of the target object, it can be difficult to evaluate color unevenness.

[0094] Specifically, when evaluating color unevenness on a highly glossy surface such as the paint of an automobile, capturing an image of the object may result in the images of objects existing around the object being captured in the captured image.

[0095] In addition, when evaluating color unevenness using a three-dimensional object as the target, illuminance unevenness may occur in the captured image depending on the surface shape of the three-dimensional object. Since the difference between the original color and the color of the uneven color is very small, the above-mentioned reflection or illuminance unevenness makes it difficult to evaluate the color unevenness.

[0096] If the imaging range of the object captured by the imaging unit is reduced in order to suppress the effects of the above-mentioned reflections or uneven illumination, it becomes necessary to capture images multiple times in order to evaluate the entire object, which lengthens the evaluation time.

[0097] Patent Document 1 also discloses a configuration that includes an imaging unit having an area camera, a number of planar lighting devices arranged at a certain angle with the axial direction of the area camera, and a diffusion plate arranged around the lens of the area camera for diffusing the illumination light from the planar lighting devices, and an articulated robot having a function of controlling the position and angle of the imaging unit relative to an object to be imaged, and that detects defects in the object from an image obtained by irradiating the object with illumination light and imaging it.

[0098] However, in the configuration of Patent Document 1, the surface reflected light of the lighting is received strongly over the entire area to be detected. As a result, when evaluating color unevenness in dark paint such as black, the slight color difference between the original paint color and the area with color unevenness is hidden by the surface reflected light of the lighting. As a result, no difference in brightness value occurs in the captured image, and color unevenness cannot be evaluated accurately. In addition, since the area that can be detected at one time on the object is small, it takes time to evaluate color unevenness over the entire object.

[0099] The color unevenness evaluation device 100 according to this embodiment has a projection target member 3 that is projected onto the surface of the object S, and the captured image 7 includes an image 70 of the projection target member 3 projected onto the surface of the object S.

[0100] By having the projection member 3, the images of objects such as walls and indoor lighting fixtures present around the target object S can be blocked, preventing them from being projected onto the surface of the target object S, thereby preventing these objects from being reflected in the captured image 7.

[0101] Furthermore, by making the color of the projection target 3 approximately uniform, the color of the image 70 of the projection target 3 projected onto the surface of the object S becomes approximately uniform, and unevenness in the illuminance of the light that the illumination unit 1 illuminates the object S can be suppressed.

[0102] As a result, in this embodiment, the influence of images of objects around the object being projected onto the object surface and the influence of uneven illuminance, etc., can be suppressed, and color unevenness can be accurately evaluated. In addition, since there is no particular restriction on reducing the imaging range of the object S by the imaging unit 2, color unevenness can be evaluated in a short evaluation time.

[0103] Furthermore, this embodiment has an extraction unit 41 that extracts an evaluation image 71 in which an image 70 of the projection target member 3 is projected from the captured image 7. By using the evaluation image 71 extracted in this manner, it is possible to suppress the images of objects around the target object from being transferred onto the target object surface and the effects of uneven illuminance, and to accurately evaluate color unevenness.

[0104] Moreover, this embodiment has an ideal image generating unit 42 that generates an ideal image 74 of the object S without color unevenness based on the captured image 7. Also, it has a differential image acquiring unit 43 that acquires a differential image 75 that is a differential image between the captured image 7 and the ideal image 74. This makes it possible to accurately extract the differential image 75 from the captured image 7 and accurately evaluate color unevenness.

[0105] Furthermore, in this embodiment, a color unevenness evaluation value E1 is output, which is acquired based on the sum of the absolute values ​​of the pixel values ​​included in the difference image 75. By using this color unevenness evaluation value E1, color unevenness can be accurately evaluated.

[0106] Furthermore, in this embodiment, it is also possible to output a corrected color unevenness evaluation value E2 corrected according to the chromaticity value of the object S. By using the corrected color unevenness evaluation value E2, it is possible to suppress differences in the correlation between the subjective evaluation and the color unevenness evaluation value according to the color of the object S.

[0107] <Modification> Here, a modified example of the first embodiment will be described. The same components as those described in the first embodiment will be given the same part numbers, and duplicated descriptions will be omitted as appropriate. This also applies to the modified examples and embodiments described below.

[0108] 10 is a diagram showing an example of the configuration of a color unevenness evaluation device 100a according to a first modified example of embodiment 1. As shown in FIG.

[0109] The light-emitting member 5 is a surface-emitting LED light with uniform illuminance within the surface, and is an example of a projection target member projected onto the surface of the target object S. By providing the light-emitting member 5 instead of the projection target member 3 in the first embodiment and making the image captured by the imaging unit 2 include the image of the light-emitting member 5 projected onto the surface of the target object S, it is possible to obtain the same effects as in the first embodiment.

[0110] 11 is a diagram showing an example of the configuration of a color unevenness evaluation device 100b according to a second modified example of embodiment 1. As shown in FIG.

[0111] The projection unit 6 has a projector 61 and a diffusion plate 62. The projector 61 projects light having a substantially uniform color toward the diffusion plate 62. The diffusion plate 62 is a plate-shaped member containing a material such as resin or glass, and diffuses the light projected by the projector 61. The projection unit 6 is an example of a member to be projected onto the surface of the target object S.

[0112] By providing a projection unit 6 instead of the projection target member 3 in the first embodiment and making the image captured by the imaging unit 2 include an image of the diffusion plate 62 projected onto the surface of the target object S, it is possible to obtain the same effect as in the first embodiment.

[0113] However, if the light emission intensity by the light emitting member 5 or the projection unit 6 is high, the shading of color unevenness may be buried in the light intensity of the reflected light on the surface of the target object S, and the color unevenness component may not be accurately extracted from the captured image. Therefore, it is preferable that the light emission intensity by the light emitting member 5 or the projection unit 6 is low.

[0114] In the above-described first embodiment, the configuration in which the extraction unit 41 extracts the evaluation image 71 including the image 70 of the light emitting member 5 reflected on the surface of the target object S from the captured image by trimming has been exemplified, but the present invention is not limited to this. The extraction unit 41 may be configured to automatically extract an appropriate area corresponding to the evaluation image 71.

[0115] Fig. 12 is a diagram for explaining a modified example of the process by the extraction unit 41. Fig. 12(a) is a diagram showing an image before the process, and Fig. 12(b) is a diagram showing an image after the process.

[0116] When the projection target 3 is white, as shown in FIG. 12(a), an image region U in which the image of the projection target 3 is reflected on the surface of the object S is brighter than an image region U' in which the image is not reflected.

[0117] Taking advantage of this, the extraction unit 41 converts the captured image 7 into an image expressed in L*, and then extracts the image area where L* is equal to or greater than a predetermined threshold, thereby automatically extracting the image area in which the projection target 3 is reflected. As a method for setting the threshold, general binary image processing methods such as the mode method and discriminant analysis method can be applied.

[0118] Next, as shown in FIG. 12(b), the extraction unit 41 extracts an evaluation image 71'' by replacing pixel values ​​included in image areas in the captured image 7 other than the image area in which the image of the projection member 3 is reflected with 0. Thereafter, the ideal image generation unit 42 performs fitting only within the evaluation image 71'' to generate an ideal image.

[0119] When the difference image acquisition unit 43 acquires a difference image between the original image based on the evaluation image 71'' and the ideal image, the pixel values ​​outside the reflected area are always 0, so that color unevenness can be evaluated only within the reflected area.

[0120] In the first embodiment, the RGB image is converted to the L*a*b* space to improve the evaluation accuracy, but it is also possible to obtain the color unevenness evaluation value directly from the RGB image before conversion. For example, it is also possible to substitute the G image for L* in equation (4). This reduces the calculation processing load.

[0121] In the first embodiment, a color area camera is used as the imaging unit 2, but the present invention is not limited to this. For example, the imaging unit 2 may include a monochrome camera, and color unevenness may be evaluated using a grayscale image acquired by the monochrome camera. Also, a multispectral camera or a hyperspectral camera may be used to acquire and evaluate an L* image from the obtained spectral image.

[0122] [Second embodiment] Next, a color unevenness evaluation device 100c according to a second embodiment will be described. In this embodiment, a method for enlarging the area on the surface of the object S onto which the projection target 3 is reflected will be described.

[0123] Fig. 13 is a diagram showing an example of the configuration of a color unevenness evaluation apparatus 100c. As shown in Fig. 13, the color unevenness evaluation apparatus 100c has a projection target member 3c. The projection target member 3c is a member that includes a curved surface.

[0124] Here, when the projection target 3 is planar as in the first embodiment, the area onto which the image of the projection target 3 is reflected may become small depending on the shape of the object S. In particular, when the curvature of the object S is large, the reflected area tends to become small.

[0125] In order to expand the projected area, it is conceivable to make the shape of the projection target member 3 as close as possible to the shape of the object S, but it is difficult to make the entire projection target member 3 close to the object S, which has a large curvature.

[0126] Therefore, in this embodiment, the shape of the projection target 3c is made to cover the object S, so that the entire projection target 3 is made to approximate the shape of the object S. This makes it possible to expand the area on the surface of the object S onto which the image of the projection target 3c is projected. Note that the shape of the projection target 3c may include a hemispherical curved shape such as a dome, may be a shape combining multiple flat surfaces, or may be a shape combining a curved surface and multiple flat surfaces.

[0127] Effects other than those described above are similar to those described in the first embodiment. [Third embodiment] Next, a color unevenness evaluation device 100d according to a third embodiment will be described. In this embodiment, a method for evaluating color unevenness in a large area of ​​the object S will be described.

[0128] 14 is a diagram showing a first example of the configuration of a color unevenness evaluation device 100d. As shown in FIG.

[0129] The holding unit 8 is a member that holds the illumination unit 1, the imaging unit 2, and the projection target member 3. The holding unit 8 is a plate-shaped member that contains metal or resin, and holds the illumination unit 1, the imaging unit 2, and the projection target member 3 by fixing them to the front surface with adhesive, screws, or the like.

[0130] However, the holding unit 8 is not limited to a plate-shaped member, and may be composed of a box-shaped member and hold the illumination unit 1, the imaging unit 2, and the projection target member 3 inside the box-shaped member. In addition, it is sufficient for the holding unit 8 to be able to hold at least the imaging unit 2 and the projection target member 3.

[0131] The moving unit 9 changes the relative positions of the illumination unit 1, the imaging unit 2, and the projection target member 3 with respect to the object S by moving the holding unit 8 along a holding unit moving direction 10. The moving unit 9 can be configured to include, for example, a linear stage and a drive unit such as a motor that drives the linear stage. The moving unit 9 may also be configured to include a robot arm or the like.

[0132] Here, in the first or second embodiment, it is necessary to increase the size of the projection target member 3 in order to increase the area to be evaluated on the object S. However, when evaluating a large object such as a door panel or a fender part of an automobile body as the object S, the size of the projection target member 3 becomes large, and the color unevenness evaluation device also becomes large.

[0133] In this embodiment, the surface of the object S is imaged multiple times while changing the relative positions of the illumination unit 1, the imaging unit 2, and the projection target 3 with respect to the object S. This makes it possible to evaluate color unevenness in a large area of ​​the object S.

[0134] In this embodiment, the imaging unit 2 may be configured to include a line sensor camera, and imaging may be performed so that the illumination unit 1, imaging unit 2, and projection target member 3 scan the target object S. In this configuration, the imaging range of the line sensor camera only needs to be included in the area where the image of the projection target member 3 is captured, so the size of the projection target member 3 can be reduced.

[0135] As described above, in this embodiment, color unevenness in a large area of ​​the object S can be evaluated by the small-sized color unevenness evaluation device 100d.

[0136] 14 illustrates an example of a configuration in which the holder 8 is moved, but the holder 8 may be fixed and the object S may be moved to change the relative positions of the illumination unit 1, the imaging unit 2, and the projection target member 3 with respect to the object S. Fig. 15 is a diagram showing a second example of the configuration of such a color unevenness evaluation device 100e.

[0137] 15, the color unevenness evaluation device 100e has a moving unit 9e. The moving unit 9e moves the object S along an object moving direction 10e. The moving unit 9e can be configured to include a linear stage and a driving unit such as a motor, or can be configured to include a robot arm or the like. Also, a belt conveyor for transporting the object S in a production line can be used as the moving unit 9e as it is.

[0138] Although the embodiments have been described above, the present invention is not limited to the above specifically disclosed embodiments, and various modifications and changes are possible without departing from the scope of the claims.

[0139] For example, the embodiment includes a color unevenness evaluation system having at least one of the color unevenness evaluation devices 100, 100a, 100b, 100c, 100d, and 100e described above, and an information processing device.

[0140] The information processing device is communicatively connected to at least one of the color unevenness evaluation devices 100, 100a, 100b, 100c, 100d, or 100e, and is capable of processing the received information relating to color unevenness of the object S. The information processing device can be constructed by a computer such as a cloud server, for example.

[0141] In addition, all numbers such as ordinal numbers and quantities used in the description of the embodiments are merely examples for specifically explaining the technology of the present invention, and the present invention is not limited to the exemplified numbers. Furthermore, the connection relationships between the components are merely examples for specifically explaining the technology of the present invention, and the connection relationships for realizing the functions of the present invention are not limited to these.

[0142] The embodiment also includes a color unevenness evaluation method. For example, the color unevenness evaluation method includes a step of outputting a color unevenness evaluation value of an object acquired based on a captured image including an image of a projection target member projected onto a surface of the object. Such a color unevenness evaluation method can provide the same effect as the above-mentioned color unevenness evaluation device.

[0143] The embodiment also includes a program. For example, the program causes a computer to execute a process of outputting a color unevenness evaluation value of an object acquired based on a captured image including an image of a projection target member projected onto a surface of the object. By using such a program, it is possible to obtain the same effect as the above-mentioned color unevenness evaluation device.

[0144] Furthermore, each function of the above-described embodiments can be realized by one or more processing circuits. Here, the term "processing circuit" in this specification includes a processor programmed to execute each function by software, such as a processor implemented by an electronic circuit, and a device such as an ASIC (Application Specific Integrated Circuit), a DSP (digital signal processor), an FPGA (field programmable gate array), or a conventional circuit module designed to execute each function described above. [Explanation of symbols]

[0145] 1. Lighting section 2. Imaging unit 3 Projected member 4 Processing section 41 Extraction part 42 Ideal image generation section 43 Differential image acquisition unit 44 Evaluation Section 45 Output section 5. Light-emitting component (an example of a projected component) 6 Projection unit (an example of a projection target) 61 Projector 62 Diffuser 7. Captured images 70 statues 71 Evaluation Images 73 Original image 74 Ideal Image 75 Difference Images 8 Holding part 9. Moving Part E1 color unevenness evaluation value E2 correction color unevenness evaluation value M Uneven color θ1 Illumination angle θ2 Imaging angle θ3 Projection angle U, U' image area [Prior art documents] [Patent documents]

[0146] [Patent Document 1] JP 2011-75534 A

Claims

1. An imaging unit that captures an image of an object; A projection member that is projected onto a surface of the object; a processing unit that outputs a color unevenness evaluation value of the object, The processing unit includes: generating an ideal image of the object without color unevenness based on the captured image including an image of the projection target projected onto a surface of the object; a color unevenness evaluation device that outputs the color unevenness evaluation value obtained based on the sum of absolute values ​​of pixel values ​​included in a difference image between the captured image and the ideal image.

2. A color unevenness evaluation device as described in Claim 1, wherein the processing unit generates the ideal image based on an evaluation image that reflects an image of the projected member, extracted from the captured image.

3. The color unevenness evaluation device according to claim 1 , wherein the processing unit outputs the color unevenness evaluation value corrected in accordance with a chromaticity value of the object.

4. The color unevenness evaluation device according to claim 1 , wherein the projection target includes at least one of a curved surface and a plurality of flat surfaces.

5. A holding unit that holds the imaging unit and the projection target member; a moving unit that changes a relative position between the holding unit and the object, 5 . The color unevenness evaluation device according to claim 1 , wherein the imaging section images the surface of the object a plurality of times while the relative position of the object is changed by the moving section, to obtain the captured images.

6. The color unevenness evaluation device according to claim 1 , a color unevenness evaluation system including: an information processing device communicably connected to the color unevenness evaluation device and capable of processing information relating to color unevenness of the object.

7. A step of acquiring an image of an object by an imaging unit; and outputting a color unevenness evaluation value of the object by a processing unit, The processing unit includes: generating an ideal image of the object without color unevenness based on the captured image including an image of the projection target member projected onto the surface of the object; The color unevenness evaluation method outputs the color unevenness evaluation value obtained based on the sum of absolute values ​​of pixel values ​​included in a difference image between the captured image and the ideal image.

8. An imaging unit is caused to acquire an image of an object, The processing unit includes: generating an ideal image of the object without color unevenness based on the captured image including an image of the projection target member projected onto the surface of the object; A color unevenness evaluation value of the object is output based on a sum of absolute values ​​of pixel values ​​included in a difference image between the captured image and the ideal image. A program that causes a computer to carry out processing.

Images