JP2000009534A - Color prediction method, color identification method and system therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate color of a product in an actual environment, based on three stimulation values XYZ of the product. SOLUTION: Mutual conversions I-VIII are determined between three stimulation values 12 of a sample of a specified material, three stimulation values 14 of a color chip seen in the same color, an output 16 of a CCD camera 16 when the sample in a specified environment is photographed by the CCD camera, a monitor input 18 for outputting color seen in the same color and a printer input 20 for outputting color seen in the same color to be stored into a computer. To predict the color of the sample with the three stimulation values 12 known, the conversions from VIII to II and III or V are performed and to estimate the three stimulation values of the color outputted to a monitor, the conversions from IV to I and VII are performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and system for color evaluation, and more particularly, to a method in which a sample whose color characteristic values such as tristimulus values X, Y and Z are known is placed in a specific environment. Method and system for color prediction for predicting the color observed when the image is output to an image output device (color output device) such as a color monitor and a color printer, and image output such as a color monitor and a color printer Method and system for color identification for estimating color characteristic values such as tristimulus values X, Y, Z of a sample when the color output to the device is assumed to be the color of the sample placed in a specific environment About.

[0002]

2. Description of the Related Art Products whose color is an important factor, such as cosmetics and paints, are managed by color characteristic values such as tristimulus values X, Y, Z, and RGB values. Tristimulus values X, Y, Z
Is determined by calculation from the result of illuminating a sample placed in an ideal environment with a standard light source and measuring the reflected light with a spectrophotometer, as defined in JIS Z8722. CI between these tristimulus values X, Y, Z and the R, G, B color system
The following conversion formula is set by E (International Commission on Illumination).

[0003]

(Equation 1)

[0004] Therefore, X,
Each element of the 3 × 3 matrix is determined by measuring the Y, Z, R, G, and B values and substituting the measured values into equation (1).
It is conceivable that color identification is performed by RGB → XYZ conversion according to the equation, and color prediction is performed by XYZ → RGB conversion using an inverse matrix. In particular,
Tristimulus values X, Y, and Z are measured using a spectrophotometer for several samples placed in an ideal environment and illuminated with standard light, and an imaging device (color measuring device) such as a CCD camera is measured for the same sample. The R, G, and B values are measured to determine the matrix and inverse matrix of equation (1). In color prediction, given tristimulus values X, Y, Z are converted to R, G, B values using an inverse matrix and input to a color monitor or color printer having RGB inputs. Output the predicted color. In color identification, R, G, and B input values of a color displayed on a color monitor or a color printer are converted into tristimulus values X, Y, and Z according to the equation (1).

[0005]

However, the above-described method has the following problems. First, the matrix is determined from the measurement values obtained using the standard light in an ideal environment. However, the environment where the color of the product is actually observed is not an ideal environment, but the relative spectral distribution is originally There is no such thing as a defined standard light. The environment of a normal color evaluation is, for example, a condition as shown in FIG. 1, and it is impossible to realize an evaluation in proportion to a feeling of an evaluator without capturing the environment itself as a parameter.
That is, the above-described color evaluation method is executed in a specially set space different from the space actually perceived by each person.

More specifically, as shown in FIG.
For example, the reflectance determined by the surface roughness and absorption characteristics of the cloth material 100 to be evaluated, the color of the table 101 itself on which the material 100 is installed, the direct light 103 from the illumination light source 102 and the like, and reflection by the surroundings Reflected light 1
04 etc., the color of the material 100 is influenced.
In addition, the color sensation also changes depending on the individual evaluator's intention, gender, and the like.

FIG. 2A shows a part P of the material 100 shown in FIG.
FIG. 2B is a schematic enlarged view of FIG.
It is a typical enlarged view of a part Q of (A). As shown in FIG. 2A, light 103 and 104 from an illumination light source
Is transmitted and absorbed by the surface roughness and the absorption characteristics of the material 100, and as shown in FIG. 2B, the light transmitted from the gap of the material 100 is reflected by the table 101, and the color of the table 101 is changed. Is added to the color evaluation of the material 100.

As shown in FIG. 3, when the color of a material is projected on a color monitor, the self-luminous component 105 of the color monitor 110 and the illumination light source 102 'are used.
And the like, and the reflected component 106 of the ambient light is superimposed and enters the eyes of the evaluator, so that there is a problem that correct color evaluation is not achieved. Second, in equation (1), there is a linear relationship between XYZ values and RGB values (that is, the matrix values are RGB
(It does not depend on the B value), but there is a problem that a linear relationship is not actually established between the RGB value and the XYZ value output by an imaging device such as a CCD camera.

Third, if the materials of the samples are different, for example, if one of the samples is a material made of powder such as eyeshadow and the other is a material having a surface gloss such as lipstick, one person When the spectral distribution of the reflected light is measured even when it is determined that the colors are the same by comparison, they are different from each other, and thus different tristimulus values X, Y, and Z may be given. Conversely, even if the X, Y, and Z values are the same, the colors seen by humans may be different if the materials are different. That is, when evaluating the color of a product with a different material, it is necessary to evaluate not only the tristimulus values X, Y, and Z but also the factors of the material of the product.

Accordingly, a first object of the present invention is to evaluate the color of a product in a specified real environment based on color characteristic values such as tristimulus values X, Y and Z of the product. It is to provide possible methods and systems. A second object of the present invention is to provide a method and a system capable of evaluating colors using a conversion equation taking into account non-linearity.

A third object of the present invention is to provide a method and a system capable of evaluating a color in consideration of a factor due to a difference between materials in addition to a color characteristic value of a product.

[0012]

According to the present invention, a) a method of converting a color characteristic value of a sample into a color output value when the sample is measured by a color measuring device in a specific environment. B) determining the color output value as a color input value, wherein the color output device looks substantially identical to the color of the sample in the particular environment when input to the color output device; Determining a second conversion equation for converting the color into an output one; c) for a specific sample given the color characteristic value, from the given color characteristic value according to the first conversion equation, Calculating an estimate, d) calculating an estimate of a color input from the estimate of the color output according to the second conversion equation, e) inputting the estimate of the color input to the color output device. The color when the specific sample is in the specific environment Color prediction method comprising the steps of reproducing the color output device on is provided.

Preferably, the first conversion equation determined in the step a) converts a color characteristic value into a color output value for a sample of a first material, and the method includes the steps of: And determining a third conversion formula for converting the color characteristic value of the sample of the material of the second material into the color characteristic value of the sample of the first material having a color that looks substantially the same as the color of the sample of the second material. G) calculating the color characteristic value of the sample of the first material from the color characteristic value of the sample of the second material according to the third conversion formula for use in the calculation in the step c). I do.

According to the present invention, a) the color input value is converted from a sample exhibiting, in a specific environment, a color which looks substantially the same as the color on the color output device to which the color input value is input, through a color measurement device. Determining a first conversion equation to convert to a resulting color output value; b) converting the color output value to a color characteristic value of the sample;
Determining a second conversion formula; c) inputting a specific color input value to a color output device to output a color; d) a color input value input to the color output device according to the first conversion formula. E) calculating an estimated value of the color output corresponding to the color output from the estimating unit, and e) calculating a color characteristic value corresponding to the color output from the estimated value of the color output according to the second conversion formula. A provided color identification method is also provided.

Preferably, the second conversion equation determined in the step b) converts the color output value into a color characteristic value of a sample of the first material, and the method includes the steps of: Determining a third conversion formula for converting the color characteristic value of the sample of the material into the color characteristic value of the sample of the second material having a color that looks substantially the same as the color of the sample of the first material; g) calculating the second characteristic from the color characteristic value of the sample of the first material calculated in the step e) according to the third conversion formula;
Calculating the color characteristic value of the sample of the material.

According to the present invention, means for storing a first conversion formula for converting a color characteristic value of a sample into a color output value when the sample is measured by a color measuring device in a specific environment, Converting the color output value into a color input value that, when input to the color output device, outputs a color that appears substantially identical to the color of the sample in the particular environment. Means for storing the conversion formula of (2); means for calculating an estimated value of color output from the given color property value according to the first conversion formula for a specific sample given the color property value; Means for calculating an estimated value of the color input from the estimated value of the color output in accordance with the conversion formula of 2, and inputting the estimated value of the color input to the color output device so that the specific sample is in the specific environment. The color output device Color prediction system comprises means for reproducing the above is provided.

Preferably, the first conversion equation is for converting a color characteristic value of a sample of a first material into a color output value, and the system converts the color characteristic value of a sample of a second material. Means for storing a third conversion formula for converting into a color characteristic value of the sample of the first material exhibiting a color substantially similar to the color of the sample of the second material, and an estimated value of the color output Means for calculating the color characteristic value of the sample of the first material from the color characteristic value of the sample of the second material according to the third conversion formula.

According to the present invention, a color input value can be obtained through a color measurement device from a sample exhibiting, in a specific environment, a color which looks substantially the same as the color on a color output device to which the color input value has been input. Means for storing a first conversion formula for converting the color output value into a color output value; means for storing a second conversion formula for converting the color output value into a color characteristic value of the sample; Means for inputting to the output device to output a color, and means for calculating an estimated value of a color output corresponding to the color output from the color input value input to the color output device according to the first conversion formula. A color identification system comprising means for calculating a color characteristic value corresponding to the color output from the color output estimate according to the second conversion equation.

Preferably, the second conversion formula converts a color output value into a color characteristic value of a sample of the first material, and the system converts the color characteristic value of the sample of the first material into: Means for storing a third conversion formula for converting into a color characteristic value of the sample of the second material exhibiting a color which looks substantially the same as the color of the sample of the first material, and according to the third conversion formula And a means for calculating a color characteristic value of the sample of the second material from a color characteristic value of the sample of the first material calculated by the color characteristic value calculating means.

[0020]

In the color prediction, a color input value for causing a color output device to output a color that looks the same as the color of a sample in a specific environment from a color characteristic value using the first and second conversion formulas. Is determined, the color in the actual environment can be reproduced based on the color characteristic value of the sample. Since linearity is not assumed in these conversion equations, it is possible to reproduce colors in consideration of the nonlinearity of the conversion. Further, since the color characteristic value of the sample is corrected in advance by using the third conversion formula to that of the standard material, it is possible to reproduce the color in consideration of the factor due to the difference in the material of the product.

In the identification of the color, the first and second conversion equations are used to calculate the color characteristic value of the sample from the color input value of the color output device that outputs the same color as the color of the sample in a specific environment. Is determined, the color characteristic value can be determined from the color of the sample in the real environment. Since linearity is not assumed in these conversion equations, it is possible to identify colors in consideration of the nonlinearity of the conversion. Furthermore, since the color characteristic value of the sample is finally corrected to that of the standard material by the third conversion formula, it is possible to identify the color in consideration of the factor due to the difference in the material of the product.

[0022]

FIG. 4 is a conceptual diagram illustrating the basic configuration of the present invention. Block 10 represents the tristimulus values X, Y, and Z of the product measured according to JIS Z8722, and block 12 therein is the tristimulus value X 'measured for a product of a specific material such as lipstick or eyeshadow. , Y ′, Z ′, and the block 14 is a tristimulus value X, Y, Z measured for a standard material such as a color mark or a tristimulus value X ′, Y measured for a product of a different material. ', Z'
Is corrected to X, Y, Z of color markers that look the same as those. Block 16 represents R _CCD , G _CCD , and B _CCD which are RGB values of pixels corresponding to the image of the product when the product placed in the real environment is imaged by the CCD camera. Note that R is red (Red) and G is green (Green)
n) and B represent blue. Block 18 is a block diagram of the RG to be input to the color monitor in order to output on the color monitor a color that looks the same as the color of the product in the real environment.
The B input values R _CRT , G _CRT , and B _CRT are shown. Block 20 is an RGB input R _PRT , G _PRT , B to be input to the color printer in order to output from the color printer a color that looks the same as the color of the product in the real environment.
_Expressed by _PRT . Hereinafter, the mutual conversion between the blocks will be described. I. XYZ ← RGB _{CCD In} addition to the linear conversion from RGB _CCD to XYZ using a matrix similar to the above equation (1), a remainder term f _{X, Y, Z} (R _CCD , G _CCD , B _CCD ) based on nonlinearity is added. RGB
The conversion from _CCD to XYZ is expressed as follows.

[0023]

(Equation 2)

The tristimulus values X, Y, and Z are measured using a spectrophotometer for three color markers of white, blue, and red, and the CCD camera 22 is operated in an environment schematically shown in FIG.
, R _CCD , G _CCD , and B _CCD for each color target 24 are measured. In FIG. 5, as the illumination 26, a fluorescent illumination having a color temperature of about 5600K, which is used for evaluation of printed matter or the like, is preferable. The distance and angle between the illumination 26 and the color watch stand 28 are set, and appropriate shielding (such as covering the surroundings with a dark curtain) from disturbance light is performed to obtain substantially constant environmental conditions. Also,
The relationship between the illumination 26, the color check table 28, and the CCD camera 20 is designed to match the viewpoint of a human when directly viewing the color mark 24 instead of photographing with the CCD camera 22. By photographing a color mark in which a plurality of colors are displayed with a CCD camera, a plurality of colors of R _CCD , G _CCD , B
You can get a _CCD .

For three colors of white, blue and red, X, Y,
Z, R_CCD, G_CCD, B_CCDWhen the value of is obtained, (2)
With the second term on the right side of the equation being 0, these values are substituted,
Matrix [a_ij] Each element a_ij(I, j = 1, 2,
3) is calculated. Next, for other color markers,
X, Y, Z, R_CCD, G_CCD, B_{CC D}Measure the value of
Then, by substituting into the equation (2), the second term on the right side is calculated.
R_CCD, G_CCD, B_CCDFor each value combination of
Remainder term f_{X, Y, Z}(R_CCD, G_CCD, B_CCD)
Calculate. And R_CCD, G_CCD, B_CCD, F
_X(R_CCD, G _CCD, B_CCD), F_Y(R_CCD,
G_CCD, B_CCD) And f_Z(R_CCD, G_CCD,
B_CCD) Is the value of the self-proliferating neural circuit (BNN: Breeding N).
eural Network)
Title: Humanic Memory HM-2000 (Nikkei Intel
Regent System Separate Volume 1992 Summer Issue 120-127
Page and JP-A-5-282271).
To learn. This gives R_CCD, G _CCD, B_CCDof
When the set is input to the neural network, the remainder term f
_{X, Y, Z}(R_CCD, G_CCD, B_CCD) Estimates are output
Therefore, according to equation (2), RGB_CCDValue of
From these, tristimulus values X, Y, and Z are calculated. These measurements and
Processing such as calculation has a color monitor as shown in FIG.
Image processing computer having an input / output terminal 30
Data 32, a CCD camera 22, a spectrometer 34, and a
By connecting the manic memory 36, online
Is achieved in. In addition, about the color mark whose value of XYZ is known,
Is the color mark number and XYZ value from the input / output terminal 30.
input. The matrix [a_ij] To calculate
The colors of the color markers are not limited to the above, and convergence of neural networks is possible.
As long as there are, any combination of colors can be used. II. XYZ → RGB _CCD  The inverse transformation of the equation (2) is expressed as follows.

[0026]

(Equation 3)

By substituting the data of the white, blue, and red color markers obtained by the above-described method into the equation (3) where the second term on the right side is set to 0 in the same manner as described above, each element b of the matrix is obtained.
_ij is calculated. By substituting the data of the color markers of other colors into (3), the second term on the right side is calculated in the same manner as described above,
X, Y, Z, f _RCCD (X, Y, Z), f _GCCD (X, Y,
The values of Z) and f _BCCD (X, Y, Z) are input to the neural network for learning. As a result, when the values of X, Y, and Z are input to the neural network, the estimated value of the remainder term is output. Therefore, the RGB _{CCD is calculated} from the tristimulus value by the equation (3).
Is calculated. III. RGB _CCD → RGB _CRT above and taking into account the non-linearity as well, RGB _CCD → RGB
The conversion of _CRT is expressed as follows.

[0028]

(Equation 4)

With respect to the representative three color markers and the other color markers, the input / output terminal 30 (FIG. 5) is displayed so that colors that look the same as the respective color markers in the environment shown in FIG. 5 are displayed on the color monitor. 6) The input value RGB _CRT of the color monitor is increased or decreased, and when the same color is obtained, the RGB _CRT value at that time and the RGB obtained by photographing each color mark with the CCD camera 22 are obtained.
Measure the _CCD .

In this case, it is troublesome and time-consuming to search for the RGB _CRT value only by human judgment. Therefore, as shown in FIG.
It is preferable that the D cameras 22 and 38 simultaneously capture the color mark and the display on the color monitor, and perform automatic matching by judging coincidence / non-coincidence and increasing / decreasing the value of RGB _CRT by the computer 32.

Next, in the second stage, a plurality of colors around the color determined by the automatic matching are displayed as shown in FIG. By selecting a desired color, an RGB _CRT value corresponding to the RGB _CCD value of the color mark is determined. Alternatively, the operator interactively corrects the color determined by the automatic matching using a scroll bar or the like to determine the RGB _CRT values.

RGB corresponding to RGB _CCD value of each color mark
_{When the CRT} value is obtained, the elements c _ij and the remainder of the matrix of equation (4) are determined in the same manner as described in I and II. When the value of the remainder term can be ignored, the value may be set to 0. Further, the gray balance of the color monitor is appropriately adjusted, and c _ij (i ≠ j) = 0, and the matrix [c
_ij ] can be a diagonal matrix. IV. Represents the conversion of the _{RGB CCD ← RGB CRT RGB CCD ←} RGB CRT by the following equation.

[0033]

(Equation 5)

Matrix element d in equation (5)_ijAnd the remainder
The term is also the RGB of each color mark obtained above. _CCDAnd the value of
RGB corresponding_CRTCan be determined in the same way from the value of
it can.V. RGB _CCD → RGB _PRT  RGB_CCD→ RGB_PRTIs represented by the following equation.

[0035]

(Equation 6)

With respect to the representative three color markers and the other color markers, the input values of the color printer are set so that colors that look the same as the respective color markers in the environment shown in FIG. 5 are output from the color printer. _Increase or decrease RGB _PRT, and when it becomes the same color,
The RGB _CRT value and the corresponding RGB _CCD value at that time are measured. Prior to matching with the human eye, semi-automatic matching is performed by a computer with the configuration shown in FIG. 9 as in the case of the color monitor. In accordance with an instruction from the computer 32, a plurality of color samples having different RGB _PRTs are simultaneously output from the full-color printer 40, and photographed by the CCD camera 38. Among the RGB values of the images of the plurality of color samples, the RGB value corresponding to or near the RGB _CCD value of the color mark is selected, and the input RGB value of the output color sample corresponding to the RGB value is set to the RGB _PRT value corresponding to the color mark. decide. If there is no match or close one, this process is repeated by changing the RGB _PRT until a match is found.

Next, the RGB determined by the automatic matching
_PRTA plurality of color samples of colors around the value
Output, and R that matches or is close to the actual color
GB _PRTTo the final RGB_PRTAnd These measurements
From the matrix element e in the equation (6)
_ijAnd the value of the remainder term are determined.VI. RGB _CCD ← RGB _PRT  RGB_CCD← RGB_PRTIs defined as follows.

[0038]

(Equation 7)

The matrix element p _ij and the remainder term in the equation (7) are Is determined in the same manner from the measurement values obtained in the above. VII. X ′, Y ′, Z ′ ← XYZ The conversion of X ′, Y ′, Z ′ ← XYZ is defined by the following equation.

[0040]

(Equation 8)

A sample of a specific material (eye shadow, lipstick, etc.) is prepared for a plurality of colors, and in the environment shown in FIG. 5, color markers that look the same as those are searched, and tristimulus values are measured with a spectrophotometer. , X ′, Y ′,
Z ′ and X, Y, Z of the corresponding color markers are determined.
From these values, the matrix element q _ij and the remainder term are determined in the same manner as described above. VIII. Conversion of X ′, Y ′, Z ′ → XYZ X ′, Y ′, Z ′ → XYZ is defined by the following equation.

[0042]

(Equation 9)

Using the measured values obtained in VII,
Matrix element r _ijAnd the remainder term is determined
Is done.Color prediction The conversion formula obtained in this way is used by Humanic Memory.
If stored in the connected computer, tristimulus values
Predicts color when a known product is in a particular environment and
Output from a color monitor or a color printer.
Wear. That is, the tristimulus values X ′,
Y ′, Z ′ are converted into tristimulus values X, Y, Z of color markers,
From equation (3), R_CCD, G_CCD, B_CCDConvert to Mosquito
When displaying on the color monitor, R
_CRT, G_CRT, B_CRTAnd input to the color monitor.
You. When outputting from a color printer, R_CCD, G
_CCD, B_CCDIs given by R_PRT, G_PRT, B_PRT
And input to a color printer. Multiple types of materials
In some cases, the matrix of equation (9) and the remainder
Is determined, and the X, Y,
Convert to Z.Color identification Using the reverse conversion formulas, RGB_CRT→ RGB
_CCD→ XYZ → X'Y'Z '
The tristimulus values corresponding to the colors created interactively on
It can be determined for each material.

In the example described above, the tristimulus values XYZ are used as the color characteristic values. However, other color characteristic values such as RGB defined in the equation (1) can be used. As the neural network, other types of neural networks such as a back propagation type can be used in addition to the above-mentioned BNN.

[0045]

As described above, according to the present invention,
The color of a product in a real environment can be evaluated based on the color characteristic value of the product.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a real environment.

FIG. 2 is a diagram illustrating an example of a material structure.

FIG. 3 is a diagram illustrating an environment for real color reproduction.

FIG. 4 is a conceptual diagram illustrating a basic configuration of the present invention.

FIG. 5 is a diagram illustrating an example of a color evaluation environment.

FIG. 6 is a configuration diagram of a system for determining a conversion equation of XYZ ← RGB _CCD .

FIG. 7 is a diagram illustrating automatic matching for determining an RGB _CRT .

FIG. 8 is a diagram illustrating matching by human eyes.

FIG. 9 is a diagram illustrating semi-automatic matching for determining RGB _PRT .

[Explanation of symbols]

 22, 38 ... CCD camera 24 ... Color mark 26 ... Light source 28 ... Color stand 30 ... Input / output terminal 40 ... Full color printer

Claims (16)

[Claims] 1. a) determining a first conversion formula for converting a color characteristic value of a sample into a color output value when the sample is measured by a color measuring device in a specific environment; b) determining the color conversion value; Converting the output value to a color input value that, when input to the color output device, outputs a color that appears substantially identical to the color of the sample in the particular environment; C) calculating an estimated value of a color output from the given color characteristic value according to the first conversion formula for a specific sample given the color characteristic value; E) calculating an estimated value of the color input from the estimated value of the color output according to the conversion formula of 2), and e) inputting the estimated value of the color input to the color output device so that the specific sample is in the specific environment. Each step of reproducing the color at Color prediction method having a flop. 2. The first conversion formula determined in the step a) is for converting a color characteristic value into a color output value for a sample of a first material, and f) a conversion formula of a sample of a second material. Determining a third conversion formula for converting the color characteristic value into a color characteristic value of a sample of the first material exhibiting a color that looks substantially the same as the color of the sample of the second material; 2. The method according to claim 1, further comprising calculating a color characteristic value of the sample of the first material from a color characteristic value of the sample of the second material according to the third conversion formula for use in the calculation in c). the method of. 3. The color characteristic value is a tristimulus value X,
Y and Z, wherein the color measurement device is an imaging device that outputs R, G, and B output values as the color output values, and the color output device receives R, G, and B input values as the color input values. The method according to claim 1, wherein the image output device is used. 4. The first conversion equation is defined by the tristimulus values X,
The step a) is represented by the sum of a vector obtained by applying a transformation matrix to a vector of Y and Z and a residual vector as a function of the tristimulus values X, Y and Z. Tristimulus values X, Y, Z respectively
And (ii) substituting the zero vector for the remainder vector and substituting the tristimulus values X, Y, Z and the R, G, B output values for the three samples. (Iii) substituting the tristimulus values and R, G, B output values for the remaining samples and the determined transform matrix values for each of the tristimulus values X, Y, Calculating the value of the residual vector for Z, and (iv) making the neural network learn the value of the residual vector for each of the tristimulus values X, Y, and Z, thereby obtaining the arbitrary tristimulus values X, Y, and Z for the neural network. 4. The method of claim 3 including the sub-step of enabling the output of the value of the corresponding remainder vector when applied to the network. 5. A color output obtained through a color measuring device from a sample which presents, in a specific environment, a color input value in a specific environment, a color which appears substantially the same as the color on a color output device to which the color input value has been input. Determining a first conversion formula to convert to a value; b) converting the color output value to a color characteristic value of the sample;
Determining a second conversion formula; c) inputting a specific color input value to a color output device to output a color; d) a color input value input to the color output device according to the first conversion formula. E) calculating an estimated value of the color output corresponding to the color output from the estimating unit, and e) calculating a color characteristic value corresponding to the color output from the estimated value of the color output according to the second conversion formula. Color identification method to be provided. 6. The second conversion equation determined in the step b) converts a color output value into a color characteristic value of a sample of a first material, and f) a color of a sample of a first material. Determining a third conversion formula for converting the characteristic value into a color characteristic value of a sample of a second material exhibiting a color that looks substantially the same as the color of the sample of the first material; From the color characteristic values of the sample of the first material calculated in the step e), the second
6. The method according to claim 5, further comprising calculating a color characteristic value of the sample of the material. 7. The color characteristic value is a tristimulus value X,
Y and Z, wherein the color measurement device is an imaging device that outputs R, G, and B output values as the color output values, and the color output device receives R, G, and B input values as the color input values. The method according to claim 5, wherein the image output device is an image output device. 8. The second conversion formula is represented by a sum of a value obtained by applying a conversion matrix to the R, G, B output value vector and a remainder vector as a function of the R, G, B output value. Step a) includes: (i) tristimulus values X, Y, and Z for a plurality of samples, respectively.
And (ii) substituting the zero vector for the remainder vector and substituting the tristimulus values X, Y, Z and the R, G, B output values for the three samples. (Iii) substituting the tristimulus values and the R, G, B output values for the remaining samples and the determined values of the conversion matrix for each of the R, G, B output values. (Iv) by learning the value of the residual vector for each of the R, G, B output values in the neural network,
8. The method of claim 7, including the sub-step of enabling output of a value of a corresponding remainder vector when a B output value is provided to the neural network. 9. A means for storing a first conversion formula for converting a color characteristic value of a sample into a color output value when the sample is measured by a color measuring device in a specific environment; Into a color input value that, when input to the color output device, outputs a color that appears substantially identical to the color of the sample in the particular environment. Means for storing a conversion formula; means for calculating a color output estimated value from a given color property value in accordance with the first conversion formula for a specific sample given the color property value; Means for calculating an estimated value of the color input from the estimated value of the color output according to the conversion formula; and inputting the estimated value of the color input to the color output device, whereby the specific sample is in the specific environment. The current color is reproduced on the color output device. Color prediction system comprises means for. 10. The first conversion formula is for converting a color characteristic value of a sample of a first material into a color output value, and converting the color characteristic value of a sample of a second material into the second material. Means for storing a third conversion formula for converting into a color characteristic value of the sample of the first material exhibiting a color that looks substantially the same as the color of the sample of the material, and used for calculating the estimated value of the color output 10. The system according to claim 9, further comprising means for calculating a color characteristic value of the sample of the first material from a color characteristic value of the sample of the second material according to the third conversion formula. 11. The color characteristic value is a tristimulus value X,
Y and Z, wherein the color measurement device is an imaging device that outputs R, G, and B output values as the color output values, and the color output device receives R, G, and B input values as the color input values. The system according to claim 9, wherein the system is an image output device. 12. The first conversion equation is defined as the tristimulus value X,
The first conversion equation is represented by the sum of a vector obtained by applying a conversion matrix to a vector of Y and Z and a residual vector as a function of the tristimulus values X, Y and Z. Tristimulus values X, Y, Z for the sample
And (ii) substituting the zero vector for the remainder vector and substituting the tristimulus values X, Y, Z and the R, G, B output values for the three samples. (Iii) substituting the tristimulus values and R, G, B output values for the remaining samples and the determined transform matrix values for each of the tristimulus values X, Y, Calculating the value of the residual vector for Z, and (iv) making the neural network learn the value of the residual vector for each of the tristimulus values X, Y, and Z, thereby obtaining the arbitrary tristimulus values X, Y, and Z. 12. The system of claim 11, wherein the system is determined by enabling output of a value of a corresponding remainder vector when applied to a network. 13. A color input value converted into a color output value obtained through a color measurement device from a sample presenting in a specific environment a color which looks substantially the same as the color on the color output device to which the color input value has been input in a specific environment. Means for converting, storing a first conversion formula, and converting the color output value to a color characteristic value of the sample;
Means for inputting a specific color input value to a color output device and outputting a color; and outputting from a color input value input to the color output device according to the first conversion formula. A means for calculating an estimated value of a color output corresponding to the obtained color; and a means for calculating a color characteristic value corresponding to the color output from the estimated value of the color output according to the second conversion formula. Identification system. 14. The color conversion method according to claim 2, wherein
And converting the color characteristic value of the sample of the first material into a color that looks substantially the same as the color of the sample of the first material. Means for storing a third conversion formula for converting the color characteristic value of the sample into a color characteristic value of the sample; 14. The system according to claim 13, further comprising means for calculating a color characteristic value of the sample of the second material. 15. The color characteristic value is a tristimulus value X,
Y and Z, wherein the color measurement device is an imaging device that outputs R, G, and B output values as the color output values, and the color output device receives R, G, and B input values as the color input values. The system according to claim 13, wherein the system is an image output device. 16. The second conversion formula is expressed by a sum of a value obtained by applying a conversion matrix to the vector of the R, G, B output values and a remainder vector as a function of the R, G, B output values. The second conversion equation is as follows: (i) tristimulus values X, Y, and Z for a plurality of samples, respectively.
And (ii) substituting the zero vector for the remainder vector and substituting the tristimulus values X, Y, Z and the R, G, B output values for the three samples. (Iii) substituting the tristimulus values and the R, G, B output values for the remaining samples and the determined values of the conversion matrix for each of the R, G, B output values. (Iv) by learning the value of the residual vector for each of the R, G, B output values in the neural network,
16. The system of claim 15, wherein the B output value is determined by enabling the output of the value of the corresponding remainder vector when applied to the neural network.