JP7308443B2 - Evaluation device, authenticity evaluation method and program - Google Patents

Description

The present invention relates to an evaluation device, a real feeling evaluation method, and a program.

Conventionally, artificial or synthetic leather (hereinafter referred to as "synthetic leather") imitating the texture of natural leather has been used for inexpensive clothing, shoes and the like. In addition, due to the recent surge in the price of natural leather, synthetic leather has come to be used even in products that have traditionally used natural leather. Therefore, synthetic leather is required to have a texture closer to that of natural leather.

Japanese Patent Laid-Open No. 2002-200000 discloses a technique for evaluating the texture of "irritation" caused by resin lumps on the surface of suede-like synthetic leather based on an image feature amount for the purpose of enhancing the texture of the synthetic leather.

However, the texture of synthetic leather is judged by the qualitative “likeness” that humans perceive when looking at natural leather, but the characteristics that make up the natural leather-likeness that humans perceive are still unclear. is not. The qualitative “likeness” that humans feel when looking at natural leather is called “genuine feeling”.

The present invention has been made in view of the above. The purpose is to evaluate

In order to solve the above-mentioned problems and achieve the object, the present invention provides an evaluation device for evaluating the authenticity of the leather based on an image of the surface of the leather , wherein the chromaticity distribution in the captured image of the leather is and a statistic calculation means for calculating a statistic based on the spatial frequency characteristic based on the variation amount of the surface texture of the leather as grain or unevenness in the captured image of the leather , at least 2 according to the surface texture of the leather Divided into one or more frequency bands and integrated, using divided integral value obtaining means for obtaining respective integral values, and coefficients determined based on a plurality of subjective evaluation points obtained by subjective evaluation experiments for authenticity, Evaluation value calculation means for constructing an evaluation formula for the authenticity of the leather from the statistic and the integrated value to calculate an evaluation value indicating the authenticity of the leather ; and a surface texture acquisition means for acquiring the surface texture in the captured image of the leather, wherein the chromaticity distribution acquisition means and the surface texture acquisition means are at least two angles or more. A captured image of the leather acquired from the measurement angle condition is acquired .

ADVANTAGE OF THE INVENTION According to this invention, it is effective in the ability to quantitatively evaluate a qualitative sense of authenticity from the measured physical feature amount.

FIG. 1 is a schematic diagram showing a configuration example of an evaluation system according to an embodiment. FIG. 2 is a block diagram showing the hardware configuration of the controller. FIG. 3 is a functional block diagram showing functions of a control unit. FIG. 4 is a diagram showing an example of an apparatus for subjective evaluation experiments. FIG. 5 is a diagram showing the results of a subjective evaluation test. FIG. 6 is a flowchart schematically showing the flow of real feeling evaluation processing by the control unit. FIG. 7 is a diagram showing the relationship between the real feeling and the absolute value term. FIG. 8 is a correlation diagram between the real feeling evaluation amount (evaluation value) and the subjective evaluation points made in advance.

Embodiments of an evaluation device, a method for evaluating authenticity, and a program will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic diagram showing a configuration example of an evaluation system 100 according to an embodiment. As shown in FIG. 1, the evaluation system 100 includes a first image acquisition device 10, a second image acquisition device 20, and a controller 5. As shown in FIG.

A first image acquisition device 10 acquires an image for obtaining a feature amount relating to the surface texture of a sample. The first image acquisition device 10 includes a light source 1 as an example of a projection unit, a sample table 3, an angle changing unit 2 that varies the illumination angle of the light source 1 with respect to the sample table 3, and a camera as an example of an imaging unit. 4 and .

As the light source 1, a xenon light source having a spectral distribution close to that of sunlight is used. The light source 1 is moved by the angle changer 2 with respect to the normal L direction of the sample table 3 and positioned appropriately in a desired angular direction.

A sample O for evaluating the surface properties is attached to the sample stage 3 . The camera 4 is arranged in the normal L direction of the sample table 3 .

The camera 4 captures the sample O attached to the sample table 3 to obtain a two-dimensional image. In this embodiment, as the camera 4, a general digital camera or an RGB camera such as a CCD is used.

On the other hand, the second image acquisition device 20 acquires an image for obtaining a feature amount regarding the chromaticity distribution of the sample. The second image acquisition device 20 includes a light source 11 that is an example of a projection unit, a sample table 13, an angle changing unit 12 that varies the illumination angle of the light source 11 with respect to the sample table 13, and a camera that is an example of an imaging unit. 14 and.

As the light source 11, a xenon light source having a spectral distribution close to that of sunlight is used. The light source 11 is moved by the angle changer 2 with respect to the normal L direction of the sample table 3 and positioned appropriately in a desired angular direction.

A sample O for evaluating the chromaticity distribution is attached to the sample table 13 . The camera 14 is arranged in a direction inclined by 45° with respect to the normal line L direction of the sample stage 13 .

The camera 14 captures the sample O attached to the sample table 13 to obtain a two-dimensional image. In this embodiment, a hyperspectral camera (spectral camera) is used as the camera 14 . A hyperspectral camera is a camera that can measure the brightness of each wavelength of light, and can obtain a spectral spectrum at each pixel. The spectroscopic camera of this embodiment can acquire a wavelength range of 400 nm to 700 nm every 10 nm.

Note that the camera 14 is not limited to a hyperspectral camera, and may be a digital camera or an RGB camera such as a CCD. However, since the information of only three channels of RGB is converted to L*a*b* instead of the spectrum, the colorimetric accuracy is degraded.

Here, FIG. 2 is a block diagram showing the hardware configuration of the control section 5. As shown in FIG. As shown in FIG. 2, the control unit 5 includes a CPU 5a (Central Processing Unit) 5a, a ROM (Read Only Memory) 5b, a RAM (Random Access Memory) 5c, an HDD 5d, and the like. The control unit 5 uses the RAM 5c as a work memory to control each unit of the first image acquisition device 10, such as the light source 1, the angle changing unit 2, the camera 4, etc., according to a real feeling evaluation program pre-stored in the ROM 5b and the HDD 5d. Drive control. In addition, the control unit 5 uses the RAM 5c as a work memory according to the real feeling evaluation program pre-stored in the ROM 5b and the HDD 5d, and uses the light source 11, the angle changing unit 12, the camera 14, and the like to control the second image acquisition device 20. It drives and controls each part. As the control unit 5, for example, a personal computer (desk top, notebook computer) can be used.

The real feeling evaluation program executed by the control unit 5 of the present embodiment is a file in an installable format or an executable format, and may be configured to be recorded and provided on a computer-readable recording medium.

Furthermore, the authenticity evaluation program executed by the control unit 5 of the present embodiment may be stored on a computer connected to a network such as the Internet, and may be provided by being downloaded via the network. Also, the authenticity evaluation program executed by the control unit 5 of the present embodiment may be provided or distributed via a network such as the Internet.

The control unit 5 evaluates the surface properties and chromaticity distribution of the sample O based on the two-dimensional images captured by the cameras 4 and 14 . That is, the control unit 5 also functions as an evaluation device.

The function of the controller 5 as an evaluation device will be described with reference to the functional block diagram of FIG. As shown in FIG. 3, the CPU 5a operates in accordance with the real feeling evaluation program so that the control unit 5 includes a statistic calculation unit 51, a divided integration value acquisition unit 52, an evaluation value calculation unit 53, and a chromaticity distribution acquisition unit 54. , surface texture acquisition means 55 , spatial frequency characteristic calculation means 56 , and storage section 57 .

The chromaticity distribution acquisition means 54 acquires the chromaticity distribution in the captured image of the sample O, which is the object. In addition, the surface texture acquisition means 55 acquires the surface texture of the captured image of the sample O, which is the object. The chromaticity distribution acquisition means 54 and the surface texture acquisition means 55 acquire captured images of the sample O, which is an object, acquired under measurement angle conditions of at least two angles or more.

The statistic calculation means 51 calculates a statistic based on the chromaticity distribution in the captured image of the sample O, which is the object.

The spatial frequency characteristic calculating means 56 calculates the spatial frequency characteristic based on the variation amount of the surface texture in the captured image of the sample O, which is the object, and weights it with the visual spatial frequency characteristic. The spatial frequency characteristic calculation means 56 changes the visual spatial frequency characteristic according to the observation distance of the sample O, which is the target object, and weights the spatial frequency characteristic of the variation amount of the surface texture.

The divided integral value acquiring means 52 obtains the spatial frequency characteristic based on the variation amount of the surface texture in the captured image of the sample O, which is the object, in at least two or more frequency bands according to the surface texture of the sample O, which is the object. are divided into and integrated, and each integral value is obtained.

The evaluation value calculation means 53 uses the subjective evaluation points for the authenticity, and calculates the authenticity of the sample O as the object based on the statistic amount and the integrated value based on the chromaticity distribution in the captured image of the sample O as the object. Calculate the evaluation value shown.

Sensory evaluation information obtained in advance is stored in the storage unit 57 .

When the evaluation system 100 of the present invention evaluates the real feeling, in addition to measuring the physical quantity of the sample O, a scaled sensory evaluation score (subjective evaluation score) is required. Therefore, a subjective evaluation experiment of authenticity is performed in advance, and the sensory evaluation score (subjective evaluation score) obtained by the experiment is stored in the storage unit 57 as sensory evaluation information.

The subjective evaluation experiment will be described below. A method called paired comparison method was used for the subjective evaluation experiment. In the paired comparison method, one pair of objects to be evaluated is presented to the subject at random, and the subject is asked to score whether or not the right object feels real compared to the left object. is.

In order to conduct a subjective evaluation experiment by the Scheffe method, which is a paired comparison method, a total of 10 types of samples of natural leather and synthetic leather were created and prepared. Table 1 below shows a list of the samples used this time.

The color of the samples used this time was black, which is the most widely distributed, and the size of the samples was A4 size (length: about 300 mm, width: about 210 mm). FIG. 4 is a diagram showing an example of an apparatus for subjective evaluation experiments. As shown in FIG. 4, the texture of the leather changes depending on the angle at which it is observed, so the sample O was bent and placed on the jig 50 . The observation distance was 300 mm. Subjects compared a pair of Left Sample A and Right Sample B in the presented curved state, and based on the visually perceived authenticity of the surface of Right Sample B relative to Left Sample A, as follows: given points.

When B feels much more authentic than A +3 points When B feels more authentic than A +2 points When B feels more authentic than A +1 point A = When B 0 points When A feels more authentic than B -1 point When A feels more authentic than B -2 points A is much more authentic than B -3 points when feeling

In this experiment, the score was given in 7 levels, but other levels such as 9 levels, 11 levels, etc. may be used. As the number of steps is increased, finer differences can be detected, but the burden on the subject is increased.

The subject evaluated from a position 300 mm away. In this experiment, ₁₀ C ₂ =45 pairs were evaluated in total because there were 10 types of observation conditions. The experiment was conducted with a total of eight image evaluation engineers as subjects. The evaluation score for each sample object was calculated by applying the quantification type III analysis method to the results of the evaluation experiment described above.

Evaluation results are shown in FIG. The horizontal axis of FIG. 5 is the number (serial number) given to the sample objects, and the sample objects are arranged in the order of the subjective evaluation points. The vertical axis indicates the subjective evaluation score of each sample object, meaning that the smaller the value, the more artificial it feels. That is, the evaluation result was that the sample object "1" felt the most real, and the sample object "4" felt the most artificial.

The operation of the authenticity evaluation method executed by the evaluation system 100 configured as described above will be described below.

First, images acquired by the first image acquisition device 10 will be described.

A sample O to be evaluated is set on the sample stage 3 . The first image acquisition device 10 turns on the light source 1 , irradiates the sample O attached to the sample table 3 with light, and performs measurement with the camera 4 installed in the normal direction L of the sample table 3 . Since the texture of leather changes depending on the viewing angle, the light source 1 is moved by the angle changer 2 to illuminate the camera 4 at a highlight angle and a shade angle. Here, the highlight is a range tilted by 10 to 25° with respect to the camera 4 (the normal direction L of the sample O), and the shade is 45 to 60° with respect to the camera 4 (the normal direction L of the sample O). Indicates the tilted range. In this way, by using the highlight angle condition close to the specular reflection condition and the shade angle condition far from the specular reflection condition as the measurement angle conditions, evaluation is performed using two angle feature amounts with greater texture change due to angle change. can be done. In this embodiment, measurements at 15° and 60° were used (geometry: 15°/0°, 60°/0°). This makes it possible to consider changes in texture depending on the viewing angle. The image acquired by the first image acquisition device 10 has a measurement size of 3000×3000 pixels (imaging range: approximately 75 mm×75 mm) and a resolution of approximately 1000 dpi.

Next, images acquired by the second image acquisition device 20 will be described.

A sample O to be evaluated is set on the sample stage 13 . The second image acquisition device 20 turns on the light source 11, irradiates the sample O attached to the sample stage 13 with light, and installs the camera 14 at an angle of 45° from the normal direction L of the sample stage 13. Take measurements. Since the color of leather changes depending on the observation angle, the light source 11 is moved by the angle changer 2 to change the angle of specular reflection with the camera 14 and the angle of highlight, shade and angle in the normal direction L to the angle of specular reflection. Lighted at an angle. Here, "highlight" refers to a range tilted from 20 to 35 degrees with respect to the normal direction L of the sample O, and "shade" refers to a range tilted from 0 to -15 degrees with respect to the normal direction L of the sample O. In this way, by using the highlight angle condition close to the specular reflection condition and the shade angle condition far from the specular reflection condition as the measurement angle conditions, evaluation is performed using two angle feature amounts with greater texture change due to angle change. can be done. In this embodiment, measurements were taken at 30° for highlights and -15° for shades (geometry: 45°/45°, 30°/45°, -15°/45°). This makes it possible to consider changes in texture depending on the viewing angle. The image acquired by the second image acquisition device 20 has a measurement size of 600×600 pixels (imaging range: approximately 15 mm×15 mm) and a resolution of approximately 1000 dpi.

Next, the flow of real feeling evaluation processing executed by the control unit 5 will be described.

Here, FIG. 6 is a flow chart schematically showing the flow of real feeling evaluation processing by the control unit 5. As shown in FIG. As shown in FIG. 6, the control unit 5 acquires measurement images (captured images) using the first image acquisition device 10 and the second image acquisition device 20 (step S1).

Next, the control unit 5 converts the measured image (captured image) into an L*a*b* image in the L*a*b* color system, and acquires the chromaticity distribution (step S2). More specifically, in order to calculate L*a*b*, the control unit 5 converts an RGB image or a spectral image into an XYZ image using equations (1) and (2) shown below.

Here, S(λ) is the spectral distribution of illumination, x(λ), y(λ), and z(λ) are color matching functions, R(λ) is spectral reflectance, and k is a coefficient. k is calculated from the following equation (3). Note that the spectral reflectance was obtained by normalizing the photographed data of the sample using the photographed data of a white reference plate having a reflectance of 100% and a reflection characteristic close to perfectly diffuse radiation, and converted into a reflectance.

Then, the control unit 5 converts the XYZ image into the L*a*b* image using the following formula (4).

Here, Xn, Yn, and Zn are tristimulus values on a perfect diffuse reflection surface defined by the Commission Internationale de l'Eclairage (CIE), where Xn=96.42, Yn=100, and Zn=82.49. In this embodiment, S(λ) is the spectral distribution of D65. By converting into a color matching system that matches human senses in this way, it is possible to improve the correlation with visual evaluation.

Returning to FIG. 6, the control unit 5 (statistics calculation means 51) calculates kurtosis and skewness using the L* image among the L*a*b* images obtained by the second image acquisition device 20. Calculate (step S3). Kurtosis and skewness are a kind of statistical quantity, and are indicators of how much the population deviates from the normal distribution.

Kurtosis is a statistic that indicates how sharp a distribution is from a normal distribution, and indicates the degree of kurtosis and the spread of tails. A positive value is given when the distribution is sharper than the normal distribution, and a negative value is given when the distribution is flatter than the normal distribution (0 in the case of a normal distribution).

Skewness, on the other hand, is a statistic indicating how much the distribution is skewed from the normal distribution, and is an indicator of left-right symmetry. It takes a positive value when the right tail is long or skewed to the left, and it takes a negative value when the left tail is long or skewed to the right (0 for a symmetrical distribution).

Next, the control unit 5 (spatial frequency characteristic calculation means 56) calculates the spatial frequency characteristic using the L* image among the L*a*b* images obtained by the first image acquisition device 10 ( step S4). More specifically, the controller 5 (spatial frequency characteristic calculator 56) subtracts the L* image average value from the value of each pixel of the L* image to acquire the L* deviation image.

The deviation ΔL* of L* takes a uniformly low value in a sample such as a metal that reflects as a whole because the average value rises, and a small portion that has little effect on the average value is particularly strong locally. When the light is reflected, that is, in a so-called glittering state, the portion of the pixel that appears to shine has a large value. In other words, the human eye perceives such a pixel where the reflection is locally strong as grain or unevenness. The control unit 5 (spatial frequency characteristic calculation means 56) Fourier-transforms the acquired L* deviation image with respect to the deviation Δ in all pixels. After the Fourier transform, the control unit 5 (spatial frequency characteristic calculation means 56) averages the amplitude at each frequency to calculate a one-dimensional spatial frequency characteristic.

Next, the control unit 5 (spatial frequency characteristic calculation means 56) weights the spatial frequency characteristic obtained in step S4 with "visual spatial frequency characteristic (VTF) with respect to observation distance" (step S5). In this embodiment, the function shown in the following equation (5) is used as the visual spatial frequency characteristic. It should be noted that, as the visual spatial frequency characteristic, a function regarding other known visual characteristics may be used.

Here, the unit of ν is cycle/degree. Since the cycle/degree can be obtained by the formula (5), by multiplying the one-dimensional spatial frequency characteristic by the formula (5) calculated according to the observation distance of the sample, the texture considering the visual characteristics can be obtained. and unevenness can be evaluated. Thereby, the spatial frequency characteristic can be calculated according to the resolution of the human eye at the assumed observation distance.

Next, the control unit 5 (divided integral value obtaining means 52) divides the weighted spatial frequency characteristics obtained in step S5 into several frequency bands according to the surface properties of the leather (step S6). In this embodiment, the leather sample to be used is divided into three frequency bands. The first is 0 to 0.1 cycle/mm, which corresponds to leather unevenness. The second is 0.1 to 1 cycle/mm, which corresponds to rough grain. Finally, the third is 1 to 4 cycles/mm, which corresponds to fine texture. By doing so, it is possible to define the range of the divided frequency bands to some extent, and to calculate the integrated value corresponding to each feature of the surface texture.

Next, the control unit 5 (divided integral value acquisition means 52) performs integration as the integral value acquisition means on the frequency bands divided in step S6 (step S7).

Finally, the control unit 5 (evaluation value calculation means 53) uses the subjective evaluation score and the feature amounts calculated in steps S2, S3, and S6 to evaluate authenticity using the following equation (6). A value is calculated (step S8). Regression analysis using a stepwise method was used to calculate the evaluation formula. However, since the stepwise method outputs a model with the best coefficient of determination, it does not consider the meaning of the evaluation formula, that is, the logic for selecting explanatory variables. Therefore, we adjusted the output of the stepwise method so that it becomes a meaningful evaluation formula while maintaining statistical significance.

As shown in Table 2 below, x1, x2, x3, x4, x5, and x6 are explanatory variables, respectively, and 45°/45° L*, 60°/0° spatial frequency characteristics (0.1 to 1 cycle/ mm), -15°/45° kurtosis, 15°/0° spatial frequency characteristics (0 to 0.1 cycle/mm), 60°/0° spatial frequency characteristics (0 to 0.1 cycle/mm), 60° /0° spatial frequency characteristics (1 to 4 cycles/mm).

Also, as shown in Table 3 below, p1, p2, p3, p4, p5, p6, p7, p8, and p9 are coefficients. Each coefficient is a parameter determined based on a subjective evaluation score. Therefore, when the subjective evaluation score is calculated by changing the evaluation stage of the subjective evaluation experiment from, for example, seven stages to nine stages, the parameters change. In this experiment, as shown in Table 3 below, p1=0.3, p2=-0.02, p3=5.37, p4=-1.06, p5=-0.06, p6=0. 69, p7=13.21, p8=−0.5, p9=−0.5 (p<0.05).

なお、式（６）のパラメータは、説明変数を標準化して決定してもよい。標準化を行うことにより、単位や平均値などが異なるデータ同士を単純に比較できるようになる。

Note that the parameters of Equation (6) may be determined by standardizing explanatory variables. Standardization makes it possible to simply compare data with different units and average values.

From the equation (6), the real feeling is obtained by combining the frequency band of 0.1 to 1 cycle/mm when photographing under the shade condition, that is, the feature amount corresponding to the rough texture on the sample surface and the brightness value under the specular reflection condition. The larger the absolute value of the sum and the frequency band of 0 to 0.1 cycle/mm when shooting under highlight and shade conditions, that is, the feature amount corresponding to the unevenness on the sample surface, the larger the value. you know it will get bigger.

On the other hand, it can be seen that the value becomes smaller when the frequency band of 1 to 4 cycles/mm when photographed under shade conditions, that is, when the feature amount corresponding to the fine texture on the sample surface is large. The reason why the lightness term is given as an absolute value under the condition of rough texture and specular reflection is that subjective evaluation results show that the greater the positive or negative value of the absolute value term, the higher the score for the feeling of authenticity. is.

FIG. 7 shows the relationship between the sense of authenticity and the absolute value term. The horizontal axis of FIG. 7 is the number (serial number) given to the sample objects, and the sample objects are arranged in the order of subjective evaluation points. The vertical axis indicates the subjective evaluation score (authenticity) and the absolute value term of each sample object. The greater the positive or negative value of the absolute value term, the greater the score of the sense of authenticity.

FIG. 8 shows a correlation diagram between the authenticity evaluation amount (evaluation value) calculated by the above-described authenticity evaluation method and subjective evaluation scores obtained in advance. The contribution rate ^R2 is 0.92, and it can be seen that the evaluation amount obtained in this embodiment is extremely close to human subjectivity. As described above, according to the present invention, a method for evaluating authenticity that is highly correlated with visual observation has been developed.

As described above, according to the present embodiment, it is possible to quantitatively evaluate the authenticity (likeness of natural leather), which was previously qualitative, from the measured physical feature amount. In short, unlike the conventional evaluation method, by constructing an evaluation formula for authenticity from the physical feature values obtained by multi-angle measurement of the leather sample against the subjective evaluation score for authenticity obtained in advance, Reality, which has been qualitative until now, can be quantitatively evaluated.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the above embodiments are merely examples of the present invention, and the present invention is not limited only to the configurations of the above embodiments. . Even if there is a change in design without departing from the scope of the present invention, it is included in the present invention.

5 evaluation device 51 statistic calculation means 52 division integral value acquisition means 53 evaluation value calculation means 54 chromaticity distribution acquisition means 55 surface texture acquisition means 56 spatial frequency characteristic calculation means

JP 2017-015496 A

Claims (7)

In an evaluation device that evaluates the authenticity of the leather based on an image of the surface of the leather ,
a statistic calculation means for calculating a statistic based on the chromaticity distribution in the captured image of the leather ;
The spatial frequency characteristics based on the amount of variation in the surface texture of the leather as wrinkles or unevenness in the captured image of the leather are divided into at least two or more frequency bands according to the surface texture of the leather, and integrated. divided integral value acquisition means for obtaining an integral value;
Using coefficients respectively determined based on a plurality of subjective evaluation points obtained by subjective evaluation experiments for authenticity, constructing an evaluation formula for the authenticity of the leather from the statistic and the integrated value, evaluation value calculation means for calculating an evaluation value indicating authenticity;
Chromaticity distribution acquisition means for acquiring the chromaticity distribution in the captured image of the leather;
a surface texture acquisition means for acquiring the surface texture in the captured image of the leather;
with
The chromaticity distribution acquisition means and the surface texture acquisition means acquire captured images of the leather acquired under measurement angle conditions of at least two angles,
An evaluation device characterized by: The two or more measurement angle conditions include a highlight angle condition close to the specular reflection condition and a shade angle condition far from the specular reflection condition,
The evaluation device according to claim 1 , characterized by: Spatial frequency characteristic calculating means for calculating a spatial frequency characteristic based on the variation amount of the surface texture and weighting it with a visual spatial frequency characteristic,
The spatial frequency characteristic calculation means changes the visual spatial frequency characteristic according to the observation distance of the leather , and weights the spatial frequency characteristic of the variation amount of the surface texture.
3. The evaluation device according to claim 1 or 2 , characterized in that: The divided integral value acquisition means includes a frequency band with a longer wavelength than 0.1 cycle / mm and a frequency band with a shorter wavelength than 0.1 cycle / mm as the frequency band divided into at least two or more,
4. The evaluation device according to any one of claims 1 to 3 , characterized in that: The chromaticity distribution in the captured image of the leather and the surface texture in the captured image of the leather use L*, a*, and b* values in the L*a*b* color system.
5. The evaluation apparatus according to any one of claims 1 to 4 , characterized in that: A method for evaluating authenticity in an evaluation device for evaluating the authenticity of the leather based on an image of the surface of the leather , comprising:
A statistic calculation step of calculating a statistic based on the chromaticity distribution in the captured image of the leather ;
The spatial frequency characteristics based on the amount of variation in the surface texture of the leather as wrinkles or unevenness in the captured image of the leather are divided into at least two or more frequency bands according to the surface texture of the leather , and integrated. a divided integral value obtaining step of obtaining an integral value;
Using coefficients respectively determined based on a plurality of subjective evaluation points obtained by subjective evaluation experiments for authenticity, constructing an evaluation formula for the authenticity of the leather from the statistic and the integrated value, an evaluation value calculation step of calculating an evaluation value indicating authenticity;
A chromaticity distribution acquisition step of acquiring a chromaticity distribution in the captured image of the leather;
A surface texture acquisition step of acquiring the surface texture in the captured image of the leather;
including
The chromaticity distribution acquisition step and the surface texture acquisition step acquire captured images of the leather acquired under measurement angle conditions of at least two angles,
A genuine feeling evaluation method characterized by: A computer that controls an evaluation device that evaluates the authenticity of the leather based on an image of the surface of the leather ,
a statistic calculation means for calculating a statistic based on the chromaticity distribution in the captured image of the leather ;
The spatial frequency characteristics based on the amount of variation in the surface texture of the leather as wrinkles or unevenness in the captured image of the leather are divided into at least two or more frequency bands according to the surface texture of the leather , and integrated. divided integral value acquisition means for obtaining an integral value;
Using coefficients respectively determined based on a plurality of subjective evaluation points obtained by subjective evaluation experiments for authenticity, constructing an evaluation formula for the authenticity of the leather from the statistic and the integrated value, evaluation value calculation means for calculating an evaluation value indicating authenticity;
Chromaticity distribution acquisition means for acquiring the chromaticity distribution in the captured image of the leather;
a surface texture acquisition means for acquiring the surface texture in the captured image of the leather;
function as
The chromaticity distribution acquisition means and the surface texture acquisition means acquire captured images of the leather acquired under measurement angle conditions of at least two angles,
program for.

Images