Image processing device, image processing method, and program

Abstract

To provide, when evaluating the state of a surface of an object, processing for obtaining an evaluation result correlated to subjective evaluation made by visual observation.SOLUTION: An image processing apparatus acquires picked-up image data obtained by picking up an image of an object having an illumination image occurring on its surface, based on the picked-up image data, calculates the amount of variation in the position of an edge of the illumination image and the amount of variation in the brightness of the edge of the illumination image, and based on the amount of variation in the position of the edge and the amount of variation in the brightness of the edge, calculates an evaluation value for evaluating the state of the surface of the object.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] This invention relates to an image processing technique for evaluating the surface condition of an object. [Background technology]

[0002] When attempting to smooth the surface of industrial products through painting, the paint may harden before it becomes smooth, depending on the painting conditions. This unintended hardening of the paint can result in fine irregularities known as orange peel, which can degrade the aesthetic appeal. Patent Document 1 discloses a technique for evaluating the degree of orange peel by blocking light from a light source and determining the degree of unevenness in the shadows cast on the object. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2019-211457 [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] However, the evaluation results in Patent Document 1 sometimes showed a low correlation with subjective visual evaluations.

[0005] Therefore, the present invention aims to provide a process for obtaining evaluation results that correlate with subjective visual evaluations when evaluating the surface condition of an object. [Means for solving the problem]

[0006] To solve the above problems, the image processing apparatus according to the present invention is characterized by comprising: acquisition means for acquiring captured image data obtained by imaging an object on which an illuminated image has been generated on its surface; first calculation means for calculating the amount of change in the position of the edge of the illuminated image and the amount of change in the brightness of the edge of the illuminated image based on the captured image data; and second calculation means for calculating an evaluation value for evaluating the surface state of the object based on the amount of change in the position of the edge and the amount of change in the brightness of the edge. [Effects of the Invention]

[0007] According to the present invention, when evaluating the surface condition of an object, it is possible to obtain evaluation results that correlate with subjective visual evaluations. [Brief explanation of the drawing]

[0008] [Figure 1] Diagram showing the appearance of the evaluation system [Figure 2] A diagram to explain the evaluation of yuzu skin. [Figure 3] Diagram showing the functional configuration of the image processing device. [Figure 4] A flowchart showing the processes performed by the image processing unit. [Figure 5] A flowchart showing the processes performed by the image processing unit. [Figure 6] Diagram showing an example of a user interface [Figure 7] Diagram showing the hardware configuration of the image processing device. [Figure 8] A diagram to explain the evaluation of yuzu skin. [Figure 9] Diagram showing the functional configuration of the image processing device. [Modes for carrying out the invention]

[0009] Each embodiment will be described below with reference to the drawings. Note that the following embodiments do not necessarily limit the present invention. Furthermore, not all combinations of features described in each embodiment are essential to the solution of the present invention.

[0010] [First Embodiment] Figure 2 is a diagram illustrating the evaluation of orange peel texture. Figure 2(a) shows an example of an image 201 obtained by imaging an object on which a linear illumination image 202 has been created on the surface due to light irradiation by an illumination device. The image 201 includes the region corresponding to the illumination image 202. When the inspection surface of an object has fine irregularities called orange peel texture, fluctuations in brightness can be seen near the edges of the illumination image 202. Figure 2(b) shows an example of an image 203 which is a schematic representation of the image 201. The brightness fluctuation 205 schematically shows the brightness fluctuation visible near the edges of the illumination image 204 in the image 203. The brightness fluctuation 205 can also be seen as edge jaggedness.

[0011] Figure 2(c) is a diagram illustrating the position profile and luminance profile near the edge of the illumination image 204. Region 206 indicates the region corresponding to the captured image 203. Position profile 207 is a position profile that shows the vertical position of the edge of the illumination image 204 and is generated by a known edge detection process. Approximation line 208 is an approximation line of position profile 207. As a method for evaluating the degree of orange peel texture on the surface of an object, one method is to use the amount of variation in edge position relative to the approximation line 208, i.e., the amount of variation (standard deviation) in position profile 207, as an orange peel texture evaluation value.

[0012] However, the amount of variation in the position profile 207 sometimes showed a low correlation with the subjective evaluation values ​​obtained from subjective evaluation experiments conducted on multiple painted surfaces with different painting conditions. The factors contributing to the decreased correlation between evaluation values ​​will be explained with reference to Figure 8. Figure 8 shows the luminance distribution (distribution of pixel values) from point A to point B in the captured image 203 of Figure 2(b). The luminance distribution 209 shows the luminance distribution when an inspection surface with high gloss mapping is imaged. The luminance distribution 210 shows the luminance distribution when an inspection surface with low gloss mapping is imaged. Comparing the luminance distribution 209 and the luminance distribution 210, it can be seen that the edge slope is smaller when gloss mapping is low compared to when gloss mapping is high. Due to the smaller edge slope, the luminance variation 205 in Figure 2(b) is less likely to be visually perceived as edge jaggedness. As a result, in the method that uses the amount of variation in the position profile as the orange peel evaluation value, the orange peel evaluation value for evaluation samples with low gloss mapping was sometimes abnormally large compared to the subjective evaluation value. Furthermore, glossiness is defined in JIS K7374.

[0013] Furthermore, luminance distribution 211 shows the luminance distribution when imaging an inspection surface with high specular reflectance. Luminance distribution 212 shows the luminance distribution when imaging an inspection surface with low specular reflectance. Comparing luminance distribution 211 and luminance distribution 212, it can be seen that when specular reflectance is low, the amount of light reflected from the illumination image 204 is smaller compared to when specular reflectance is high, resulting in a smaller contrast between the edge and the background area. Due to the small contrast between the edge and the background area, the luminance fluctuation 205 in Figure 2(b) is less likely to be visible as edge jaggedness. As a result, in a method that uses the amount of fluctuation in the position profile as the orange peel evaluation value, the orange peel evaluation value for evaluation samples with low specular reflectance sometimes became abnormally large compared to the subjective evaluation value. Specular reflectance is defined in JIS Z 8741.

[0014] Furthermore, luminance distribution 213 shows the luminance distribution when imaging an inspection surface with low diffuse reflectance. Luminance distribution 214 shows the luminance distribution when imaging an inspection surface with high diffuse reflectance. Comparing luminance distribution 213 and luminance distribution 214, it can be seen that when diffuse reflectance is high, the amount of light reflected from the background area of ​​the illumination image 204 is greater than when diffuse reflectance is low, resulting in a smaller contrast between the edge and the background area. Due to the small contrast between the edge and the background area, the luminance fluctuation 205 in Figure 2(b) is less likely to be visible as edge jaggedness. As a result, in a method that uses the amount of fluctuation in the position profile as the orange peel evaluation value, the orange peel evaluation value for evaluation samples with high diffuse reflectance sometimes became abnormally large compared to the subjective evaluation value. Note that diffuse reflectance is defined in JIS Z 8105.

[0015] Furthermore, subjective evaluation experiments revealed that the wider the illumination image 204, the less noticeable the edge jaggedness became. This is thought to be because the ratio of the edge jaggedness width to the width of the illumination image 204 decreases as the width of the illumination image 204 increases. In the method using the amount of variation in the position profile as the orange peel texture evaluation value, there were cases where the orange peel texture evaluation value was abnormally large compared to the subjective evaluation value, even under conditions where the width of the illumination image 204 was large.

[0016] Therefore, in this embodiment, when dealing with inspection surfaces with low gloss mapping, low specular reflectance, or high diffuse reflectance, or under conditions where the width of the illuminated image 204 is large, the contribution of the amount of variation in the position profile to the orange peel evaluation value is suppressed. Here, C is the gloss mapping of the inspection surface, and r is the specular reflectance of the inspection surface. s The diffuse reflectance of the inspection surface is r d Let W be the width of the illumination image 204 in the captured image 203. Also, let σ be the amount of variation in the position profile 207 in Figure 2(c). p Let's assume that the Yuzu skin evaluation value E is derived by equation (1). E=(C) α (r s ) β (r d ) -γ(W) -δ σ p ··· Equation (1) In Equation (1), α, β, γ, and δ are real numbers greater than 0, and are set by known optimization processing so that the correlation between the subjective evaluation value and the yuzu skin evaluation value E is maximized. Incidentally, C, r s , r d , W, σ p are multiplied by proportional coefficients for scaling, and the proportional coefficients multiplied for each are also set by known optimization processing.

[0017] The yuzu skin evaluation value in Patent Document 1 corresponds to the case where α, β, γ, and δ are 0 in Equation (1), so E = σ p is obtained. On the other hand, the yuzu skin evaluation value in the present embodiment is such that α, β, γ, and δ are real numbers greater than 0 in Equation (1). For this reason, on an inspection surface with low specular reflectivity C or on an inspection surface with low specular reflectance r s , in Equation (1), (C) α (r s ) β becomes small, so the contribution of the variation amount σ p in the position profile 207 is suppressed. Similarly, on an inspection surface with high diffuse reflectance r d or under conditions where the width W of the illumination image 204 becomes large, in Equation (1), (r d ) -γ (W) -δ becomes small, so the contribution of the variation amount σ p in the position profile 207 is suppressed.

[0018] When using Equation (1), the specular reflectivity C, the specular reflectance r s , and the diffuse reflectance r d can be obtained using known measuring instruments. Also, as the width W of the illumination image 204, the distance between the upper and lower edges is used. The upper and lower edge positions can be derived by known edge detection processing. Incidentally, instead of the specular reflectivity C, the slope of the edge shown in FIG. 8 may be used. That is, it is considered that the higher the slope of the edge obtained based on the captured image 203, the higher the specular reflectivity. Also, the specular reflectance r s and the diffuse reflectance r dAlternatively, the edge contrast shown in Figure 8 may be used. That is, the greater the edge contrast obtained based on the captured image 203, the higher the specular reflectance and the lower the diffuse reflectance is considered to be. If the edge slope is G and the edge contrast is R, the orange peel evaluation value E is derived by equation (2). E=(G) α (R) β (W) -γ σ p ...Equation (2) In equation (2), α, β, and γ are real numbers greater than 0 and are set by an optimization process, similar to equation (1). According to equation (2), for inspection surfaces with a small edge slope G and a small edge contrast R, in equation (2), (G) α (R) β Because it becomes smaller, the amount of variation σ in position profile 207 p The contribution of is suppressed.

[0019] As explained above, the orange peel texture evaluation value using equation (1) or equation (2) shows improved correlation with the subjective evaluation value compared to the conventional orange peel texture evaluation value. Furthermore, when performing a subjective evaluation of orange peel texture, not only edge jaggedness but also brightness fluctuations are visible. In particular, in equation (1) or equation (2), the amount of fluctuation σ in the position profile 207... p When the contribution of is small, evaluators in subjective evaluation experiments focus more on luminance fluctuations than on edge jaggedness. Therefore, by adding the effect of luminance fluctuations to equations (1) and (2), the correlation between the orange peel evaluation value and the subjective evaluation value is further improved. In this case, first, the pixel values ​​of the captured image 203 in Figure 2(b) are obtained along the approximation line 208 in Figure 2(c) to generate a luminance profile 215. Then, the amount of fluctuation (standard deviation) in the luminance profile 215 is used as the luminance fluctuation σ i This is added to equations (1) and (2). As a result, equations (1) and (2) become as shown in equations (3) and (4), respectively. E=(C) α (r s ) β (r d ) -γ (W) -δ σp +kσ i ...Equation (3) E=(G) α (R) β (W) -γ σ p +kσ i ...Equation (4) k is a proportionality constant. The orange peel skin evaluation value E in equations (3) and (4) is the amount of variation σ in the position profile. p and the amount of variation σ in the luminance profile i This is a weighted sum of the following. In equation (3), (C) α (r s ) β (r d ) -γ (W) -δ is σ p With the weights, k is σ i This is the weight of. In equation (4), (G) α (R) β (W) -γ is σ p With the weights, k is σ i This is the weight of σ. i The weight k is set by optimization along with other parameters such as α, β, and γ so as to maximize the correlation between the subjective evaluation value and the yuzu skin evaluation value E. According to equation (3) or equation (4) described above, a yuzu skin evaluation value with a high correlation to the subjective evaluation value can be obtained. Below, the method for deriving the yuzu skin evaluation value will be described in detail.

[0020] <Evaluation System> Figure 1 is an external view of the evaluation system in this embodiment. The evaluation system includes an image processing device 101, an illumination device 102, and an imaging device 103. The illumination device 102 is a fluorescent tube that irradiates light onto the surface of the object 104 to be evaluated, thereby generating a linear illumination image 105 on the object 104. The imaging device 103 is a camera that images the inspection surface of the object 104 on which the linear illumination image 105 has been generated. The object 104 is part of an industrial product such as a home appliance, and the surface to be evaluated is a painted surface with a metallic luster. The image processing device 101 is connected to the imaging device 103 and controls the exposure, focus, and imaging timing of the imaging device 103. The image processing device 101 also evaluates the surface condition of the object 104 based on the image data acquired by imaging. Specifically, the image processing device 101 calculates an orange peel evaluation value corresponding to the inspection surface of the object 104 and presents it to the user.

[0021] Figure 7 is a block diagram showing the hardware configuration of the image processing device 101. The image processing device 101 includes a CPU 106, ROM 107, and RAM 108. It also includes a VC (video card) 109, a general-purpose I / F (interface) 110, a SATA (serial ATA) I / F 111, and a NIC (network interface card) 112. The CPU 106 uses RAM 108 as work memory to execute the OS (operating system) and various programs stored in ROM 107, HDD (hard disk drive) 117, etc. The CPU 106 also controls each component via the system bus 113. In the flowchart described later, program code stored in ROM 107, HDD 117, etc., is loaded into RAM 108 and executed by the CPU 106. A display device 119 is connected to VC 109. Input devices 115 such as a mouse and keyboard, and an imaging device 103 are connected to the general-purpose I / F 110 via the serial bus 114. A general-purpose drive 118, which reads and writes to the HDD 117 and various recording media, is connected to the SATAI / F111 via the serial bus 116. The NIC 112 handles the input and output of information to and from external devices. The CPU 106 uses the various recording media mounted on the HDD 117 and the general-purpose drive 118 as storage locations for various data. The CPU 106 displays the UI (user interface) provided by the program on the display device 119 and receives input such as user instructions via the input device 115. The display device 119 may be a touch panel display that has the function of a touch panel that detects the position of a touch by an object such as a finger.

[0022] Figure 6 shows an example of the UI displayed on the display device 119. The image display window 601 is a window that displays the image represented by the image data specified by the user. The evaluation range specification area 602 is the area to be evaluated, specified by a rectangle within the image. The image selection button 603 is a button for the user to specify the image data corresponding to the image they want to display. The evaluation value calculation button 604 is a button that instructs the start of calculating the orange peel evaluation value for the specified area. The contribution display text box 605 displays the contribution of the amount of variation in the position profile to the orange peel evaluation value. The evaluation value display text box 606 displays the calculated orange peel evaluation value. The exit button 607 is a button that instructs the application to exit.

[0023] Figure 3 is a block diagram showing the functional configuration of the image processing device 101. The CPU 106 functions as shown in Figure 3 by reading and executing programs stored in the ROM 107 or HDD 117 using the RAM 108 as work memory. Note that not all of the following processes need to be executed by the CPU 106; the image processing device 101 may be configured so that some or all of the processes are performed by one or more processing circuits other than the CPU 106.

[0024] The image processing apparatus 101 includes an imaging control unit 301, a position profile acquisition unit 302, a position variation calculation unit 303, a brightness profile acquisition unit 304, a brightness variation calculation unit 305, a feature quantity acquisition unit 306, an evaluation value calculation unit 307, and an evaluation value output unit 308. The imaging control unit 301 controls the imaging device 103 to image the inspection surface of an object on which a line-shaped illumination image has been generated, and acquires the image data. The position profile acquisition unit 302 acquires a position profile indicating the position of the edge of the illumination image based on the acquired image data. The position variation calculation unit 303 calculates the amount of variation σ of the edge position in the acquired position profile. pThe luminance profile acquisition unit 304 acquires a luminance profile showing the amount of luminance variation at the edge of the illumination image based on the acquired image data. The luminance variation calculation unit 305 calculates the amount of edge luminance variation σ in the acquired luminance profile. i The feature acquisition unit 306 calculates the glossiness C and specular reflectance r for the object to be evaluated. s , diffuse reflectance r d The width W of the illuminated image is obtained. The evaluation value calculation unit 307 calculates an orange peel evaluation value E to evaluate the degree of orange peel texture on the surface of the object. The evaluation value output unit 308 outputs the calculated orange peel evaluation value E.

[0025] <Processing performed by the image processing unit> In this embodiment, the processing flow performed by the image processing device 101 will be explained using the flowchart in Figure 4(a). The processing shown in the flowchart in Figure 4(a) begins when the user inputs an instruction via the input device 115 and the CPU 106 receives the input instruction. Hereafter, each step (process) will be represented by adding an S before the symbol.

[0026] In S401, the imaging control unit 301 controls the imaging device 103 to image the inspection surface of an object on which a line-shaped illumination image has been generated, and acquires image data. In S402, the position profile acquisition unit 302 acquires a position profile indicating the position of the edge of the illumination image based on the acquired image data. In S403, the position variation calculation unit 303 calculates the amount of variation σ of the edge position in the acquired position profile. p Calculate.

[0027] In S404, the luminance profile acquisition unit 304 acquires a luminance profile showing the amount of luminance variation at the edge of the illumination image based on the acquired image data. In S405, the luminance variation calculation unit 305 calculates the amount of edge luminance variation σ in the acquired luminance profile. iThe glossiness C obtained by measuring the inspection surface of the object with a known measuring instrument is calculated. In S406, the feature acquisition unit 306 acquires the specular reflectance r obtained by measuring the inspection surface of the object with a known measuring instrument. In S407, the feature acquisition unit 306 acquires the specular reflectance r obtained by measuring the inspection surface of the object with a known measuring instrument. s In S408, the feature acquisition unit 306 obtains the diffuse reflectance r obtained by measuring the inspection surface of the object with a known measuring instrument. d The feature acquisition unit 306 calculates the width W of the illumination image based on the acquired image data.

[0028] In S410, the evaluation value calculation unit 307 calculates the glossiness C and specular reflectance r. s , diffuse reflectance r d Based on the width W of the illuminated image, the amount of variation σ of the edge position relative to the orange peel skin evaluation value E. p The degree of contribution is set. In this embodiment, since equation (3) is used, the degree of contribution is (C) α (r s ) β (r d ) -γ (W) -δ The following is set. In S411, the evaluation value calculation unit 307 calculates the amount of change in edge position σ p , edge brightness variation σ i Based on the contribution to the orange peel skin evaluation value E, the orange peel skin evaluation value E for the object being evaluated is calculated.

[0029] In S412, the evaluation value output unit 308 outputs the yuzu skin evaluation value E calculated in S411. Specifically, the evaluation value output unit 308 displays the calculated yuzu skin evaluation value E in the evaluation value display text box 606. In S413, the evaluation value output unit 308 outputs the contribution level calculated in S410. Specifically, it displays the calculated contribution level in the contribution level display text box 605.

[0030] As described above, the image processing device in this embodiment acquires image data obtained by imaging an object on which an illuminated image has been generated on its surface, and calculates the amount of change in the edge position and the amount of change in edge brightness of the illuminated image based on the image data. Based on the calculated amount of change in edge position and the amount of change in edge brightness, an evaluation value for evaluating the surface condition of the object is calculated. This makes it possible to obtain evaluation results that correlate with subjective visual evaluation when evaluating the surface condition of an object.

[0031] <Variation> In the embodiment described above, the imaging control unit 301 controls imaging and acquires image data, but it may also function as an image data acquisition unit that acquires image data obtained in advance from a storage device such as a ROM 107 or HDD 117.

[0032] In the embodiment described above, the processing related to the position profile in S402 and S403 was performed first, followed by the processing related to the brightness profile in S404 and S405. However, the processing is not limited to this order. For example, the processing related to the brightness profile may be performed first, followed by the processing related to the position profile, or both processes may be performed in parallel.

[0033] In the above-described embodiment, the amount of change in edge position σ relative to the orange peel evaluation value E is used as the contribution. p The contribution of was used, but the variation in edge brightness σ relative to the orange peel skin evaluation value E i You may use the contribution of either method. Alternatively, you may calculate the contributions of both methods and display them in the contribution display text box 605.

[0034] [Second Embodiment] In the first embodiment, the yuzu skin evaluation value was calculated using Equation (3). However, in this embodiment, the yuzu skin evaluation value is calculated using Equation (4). Since the configuration of the evaluation system in this embodiment is equivalent to that of the first embodiment, the description thereof is omitted. In the following, the differences between this embodiment and the first embodiment will be mainly described. For the same configurations as those in the first embodiment, the same reference numerals will be used for description.

[0035] <Processing executed by the image processing apparatus> In this embodiment, the flow of the processing executed by the image processing apparatus 101 will be described using the flowchart of FIG. 4(b). The processing shown in the flowchart of FIG. 4(b) starts when an instruction is input by the user via the input device 115 and the CPU 106 receives the input instruction.

[0036] In S414, the feature amount acquisition unit 306 calculates the edge inclination G based on the acquired captured image data. In S415, the feature amount acquisition unit 306 calculates the edge contrast R based on the acquired captured image data. In S416, the evaluation value calculation unit 307 calculates the contribution degree of the edge position variation amount σ p to the yuzu skin evaluation value E based on the width W of the illumination image, the edge inclination G, and the edge contrast R. In this embodiment, since Equation (4) is used, (G) α (R) β (W) -γ is set as the contribution degree. In S417, the evaluation value calculation unit 307 calculates the yuzu skin evaluation value E for the object to be evaluated based on the edge position variation amount σ p , the edge luminance variation amount σ i , and the contribution degree to the yuzu skin evaluation value E.

[0037] As described above, the image processing apparatus in this embodiment calculates the yuzu skin evaluation value using a contribution degree setting method different from that of the first embodiment. Thereby, when evaluating the state of the surface of an object, an evaluation result having a correlation with the subjective evaluation by visual inspection can be obtained.

[0038] [Third Embodiment] In the second embodiment, the yuzu-skin evaluation value was calculated using Equation (4). In this embodiment, the method for calculating the yuzu-skin evaluation value is switched according to the slope G of the edge, the contrast R of the edge, the width W of the illumination image, etc. As a result, the fluctuation amount σ p calculated in the position profile and the fluctuation amount σ i calculated in the luminance profile can be omitted, and thus the yuzu-skin evaluation value can be calculated with a smaller amount of calculation than in the second embodiment.

[0039] Specifically, when the slope G of the edge is greater than the threshold Th1, the contrast R of the edge is greater than the threshold Th2, and the width W of the illumination image is smaller than the threshold Th3, the yuzu-skin evaluation value E in Equation (2) is calculated based on the fluctuation amount σ p in the position profile. Otherwise, based on the fluctuation amount σ i in the luminance profile, kσ i in the second term on the right side of Equation (4) is calculated as the yuzu-skin evaluation value E. Note that predetermined values are set for the threshold Th1, the threshold Th2, and the threshold Th3, respectively.

[0040] Note that the configuration of the evaluation system in this embodiment is different from that of the first embodiment in the functional configuration of the image processing apparatus 101. In the following, the differences between this embodiment and the above-described embodiments will be mainly described. Note that the same components as those in the above-described embodiments will be described with the same reference numerals.

[0041] FIG. 9 is a block diagram showing the functional configuration of the image processing apparatus 101. The CPU 106 functions as the functional configuration shown in FIG. 9 by reading and executing the program stored in the ROM 107 or the HDD 117 using the RAM 108 as a work memory. Note that it is not necessary for all of the following-described processes to be executed by the CPU 106, and the image processing apparatus 101 may be configured such that part or all of the processes are performed by one or more processing circuits other than the CPU 106.

[0042] The image processing apparatus 101 includes an imaging control unit 301, a position profile acquisition unit 302, a position variation calculation unit 303, a brightness profile acquisition unit 304, a brightness variation calculation unit 305, a feature quantity acquisition unit 306, an evaluation value calculation unit 307, an evaluation value output unit 308, and a determination unit 901. The determination unit 901 determines the amount of variation σ in the position profile based on the edge inclination G, the edge contrast R, and the width W of the illuminated image. p And the amount of variation σ in the brightness profile i Determine which of the two will be used to calculate the Yuzu Skin evaluation value E.

[0043] <Processing performed by the image processing unit> In this embodiment, the processing flow performed by the image processing device 101 will be explained using the flowchart in Figure 5. The processing shown in the flowchart in Figure 5 starts when the user inputs an instruction via the input device 115 and the CPU 106 receives the input instruction.

[0044] In S418, the determination unit 901 determines whether the slope G of the edge is greater than the threshold Th1. If the slope G of the edge is greater than the threshold Th1, the process proceeds to S419; otherwise, the process proceeds to S404. In S419, the determination unit 901 determines whether the contrast R of the edge is greater than the threshold Th2. If the contrast R of the edge is greater than the threshold Th2, the process proceeds to S420; otherwise, the process proceeds to S404. In S420, the determination unit 901 determines whether the width W of the illuminated image is less than the threshold Th3. If the width W of the illuminated image is less than the threshold Th3, the process proceeds to S402; otherwise, the process proceeds to S404.

[0045] In S421, the evaluation value calculation unit 307 calculates the amount of change in edge position σ p Based on this, the orange peel texture evaluation value E for the object to be evaluated is calculated using equation (2). In S422, the evaluation value calculation unit 307 calculates the variation amount σ of the edge brightness. i Based on this, the second term on the right-hand side of equation (4) is kσ iThis is calculated as the yuzu skin evaluation value E. In S423, the evaluation value output unit 308 outputs the yuzu skin evaluation value E calculated in S421 or S422.

[0046] As described above, the image processing device in this embodiment switches the calculation method for the orange peel evaluation value based on the feature quantities related to the illuminated image. This makes it possible to obtain an orange peel evaluation result that correlates with subjective visual evaluation with less computation than in the embodiment described above.

[0047] [Other embodiments] In the embodiments described above, the cases using equation (3) and equation (4) were explained separately, but the orange peel evaluation value E may also be calculated by combining equations (3) and (4). For example, in equation (4), glossiness C may be used instead of the edge slope G, and the variation in edge position σ with respect to the orange peel evaluation value E may be calculated. p The degree of contribution of (C) α (R) β (W) -γ That's also acceptable.

[0048] Furthermore, in the above-described embodiment, the variation amount of edge brightness σ i In this example, the amount of variation in the luminance profile 215 was used, but the amount of variation in luminance in a predetermined region near the edge may also be used. For example, the amount of variation in luminance (standard deviation) in a region of a predetermined width along the approximation line 208 may be used.

[0049] Furthermore, the evaluation targets and evaluation conditions are limited, and the amount of change in edge position σ relative to the orange peel skin evaluation value E is also limited. p If the contribution is always small, then the variation in edge position σ relative to the orange peel skin evaluation value E. p The contribution of k may be set to a constant. For example, as the yuzu skin evaluation value E, k p σ p +k i σ i You can output k p σ p and k i σ iThe larger of the two values ​​can be output as the Yuzu Skin Evaluation Value E. Here, k p , k i This is set through an optimization process to maximize the correlation with the subjective evaluation value.

[0050] In addition, k p , k i The k can be optimized according to the shape of the object being evaluated and the geometric conditions during imaging, and switched during evaluation. For example, k can be optimized according to the curvature of the inspection surface. p , k i The system switches between modes, and depending on the distance between the imaging device 103 and the inspection surface, and the angle of incidence of light emitted from the illumination device 102, k p , k i You can also switch between them. For example, if the inspection surface is convex or concave, the amount of variation in edge position σ will be due to the complex curve of the edges. p The value of can become abnormally large. Therefore, if the curvature of the approximation line 208 is greater than the predetermined value, k p It may be acceptable to suppress it.

[0051] Furthermore, by using the orange peel evaluation value E output by the above-described embodiment for detecting defective products, it becomes possible to detect defective products in a way that corresponds to human subjectivity. For example, the orange peel evaluation value E may be compared with a predetermined threshold, and if the orange peel evaluation value E is greater than the predetermined threshold, the system may notify that a defective product has occurred.

[0052] The present invention can also be realized by supplying a program that implements one or more of the functions of the above-described embodiments to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. It can also be realized by a circuit (e.g., an ASIC) that implements one or more functions. [Explanation of symbols]

[0053] 101 Image Processing Device 301 Imaging Control Unit 303 Position Fluctuation Calculation Unit 305 Brightness Fluctuation Calculation Unit 307 Evaluation Value Calculation Unit

Claims

1. Acquisition means for acquiring image data obtained by imaging an object on which an illuminated image has been generated on its surface, A first calculation means that calculates the amount of variation in the position of the edge of the illumination image and the amount of variation in the brightness of the edge of the illumination image based on the captured image data, A second calculation means for calculating an evaluation value for evaluating the surface state of the object based on the amount of variation in the position of the edge and the amount of variation in the brightness of the edge, An image processing apparatus characterized by having

2. The image processing apparatus according to claim 1, characterized in that the first calculation means calculates the amount of variation in the position of the edge and the amount of variation in the brightness of the edge based on a first profile indicating the position of the edge of the illuminated image and a second profile indicating the brightness of the edge of the illuminated image.

3. The image processing apparatus according to claim 1, characterized in that the second calculation means calculates the evaluation value as a weighted sum of the amount of change in the position of the edge and the amount of change in the brightness of the edge.

4. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the variation in the edge position to the evaluation value as the glossiness of the object decreases.

5. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the variation in the edge position to the evaluation value as the specular reflectance of the object decreases.

6. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the variation in the edge position to the evaluation value as the diffuse reflectance of the object increases.

7. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the amount of change in the position of the edge to the evaluation value as the inclination of the edge decreases.

8. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the variation in the position of the edge to the evaluation value as the contrast of the edge decreases.

9. The image processing apparatus according to claim 1, characterized in that the second calculation means reduces the contribution of the variation in the edge position to the evaluation value as the width of the illuminated image increases.

10. A program for causing a computer to function as one of the means of an image processing apparatus according to any one of claims 1 to 9.

11. An acquisition process to obtain image data obtained by imaging an object on which an illuminated image has been generated on its surface, A first calculation step, based on the captured image data, calculates the amount of variation in the position of the edge of the illumination image and the amount of variation in the brightness of the edge of the illumination image. A second calculation step of calculating an evaluation value for evaluating the surface state of the object based on the amount of variation in the position of the edge and the amount of variation in the brightness of the edge, An image processing method characterized by having the following features.

Images