JPH07248258A - Isochromatic value calculation method - Google Patents

Abstract

PURPOSE:To achieve efficiency in color management by performing numerical conversion of color by colorimetry to accomplish work of converting color regarded as the same based on a colorimetric value into a signal value in lump with higher reproducibility and accuracy. CONSTITUTION:As for a plurality of fiber products, a plurality of parts undergo a colorimetry for each single fiber product and a corresponding colorimetric value is stored into a color number data base. Thereafter, the following procedures (1)-(3) are carried out until none determined to be isochromatic exists to consolidate the colorimetric values: (1) a color difference is calculated about all combinations of the colorimetric values, (2) the color difference is compared with a threshold and (3) when color is determined, a mean of colorimetric values thereof is stored into the color number base anew and the previous colorimetric values are deleted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of calculating colors that can be regarded as the same color for textile products such as yarn and cloth.
More specifically, the present invention relates to a method of calculating a representative value of a color range that can be regarded as the same color, in consideration of measurement variations and manufacturing variations.

[0002]

2. Description of the Related Art In the dyeing process, the hue ordered by the user and the hue actually dyed are rarely completely the same, and color matching is performed within a certain allowable range. Therefore, if some of the dyed products in the past fall within the acceptable range of the ordered color, the resipe can be reused, and it is judged whether the target color is included in such an acceptable range. That is, it is necessary in the dyeing process to judge the same color.

Conventionally, it has been the mainstream that such a judgment of the same color is made visually by a skilled person. On the other hand, in order to make the judgment more objectively, the same color judgment method based on the measured data is considered, and the magazine Tex published in September 1992.
tile Chemistand Colorist
Vol. 24, No. On page 88 of 9, two types of shade sorting are disclosed.

One of them is the Clemson Un shown in FIG.
CCCShade Sor created by Iversity
It is a ting. This principle is, A to the color difference on a coordinate of two axes of the hue difference component [Delta] H / T _H and chroma difference components [Delta] C / T _C
Plot each batch of K. Next, batches having a small color difference are drawn in order and grouped. Then, finally, the triangles 201 and the pentagons 202 in FIG. 2 are grouped into two groups. Samples D, E, and K are determined to have the same color in the 201 group, and B and 202 in the 202 group.
It is determined that C, F, G, H, and J have the same color. A and I do not belong to any group.

The other is Coats &
555 Shad devised by Clark Color Team
e Sorting. This is also A on the color difference coordinates
Plot each batch of ~ K, set the center 301 of the color difference coordinates to 0, and subdivide the two axes into spans of 0.5 to form a grid 302.
The square formed by (dashed line) is a color group. The batches plotted in that color group were judged to have the same color. In FIG. 3, A and B and C, D and I and E
And F and H, G and J are grouped into four groups, and K does not belong anywhere.

[0006]

However, in the above visual judgment, there is a problem that reproducibility is poor when the judgment standard is not clearly written and human error is included. Furthermore, in CCC Shade Sorting, the group formation process and the final result differ depending on how the threshold is set. Therefore, the color difference between H and E in another group in FIG. There is a problem that it is smaller than the color difference.

Also, the 555 Shade Sortin
In the case of g, there is a problem that the color difference may be out of the allowable range even between batches in the same color group as in E and F in FIG.

The present invention has been made to solve the above-mentioned problems, and its purpose is to digitize colors by colorimetry and collect colors that can be regarded as the same color by the colorimetric values. It is to improve the efficiency of color management by performing the work of converting to the value of with high reproducibility and accuracy.

[0009]

The present invention provides a plurality of textile products by colorimetrically measuring a plurality of parts for each single textile product,
After the colorimetric values are stored in the color number database, the colorimetric values are integrated in the color number database by the following procedure. (A) For all combinations of arbitrary two average colorimetric values stored in the color number database after averaging the colorimetric values for each textile product to obtain the average colorimetric value and storing it in the color number database Calculate the color difference. (B) Two average colorimetric values with the smallest color difference are selected. (C) The color difference between the two selected average colorimetric values is compared with a predetermined threshold value to determine whether the colors are the same. (D) When it is determined that the colors are the same, a value obtained by averaging two average colorimetric values is newly stored in the color number database, and
The two selected average colorimetric values are deleted from the color number database, and the procedure returns to step (a). (E) If it is determined that the colors are not the same, the average colorimetric value stored in the color number database is used as the representative value of the same color range.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a flow chart showing an example of the method of the present invention, showing a color integration procedure for gathering comrades that can be regarded as the same color from a huge number of color number samples in the past.

In the procedure, a first module 11 for obtaining a threshold value from colorimetric variations and dyeing reproducibility, a second module 12 for making the same color determination using the obtained threshold value,
Further, the third module 13 performs the same color determination by the significant difference determination. In this embodiment, all colorimetric data are L ^* (lightness index), a defined by CIE.
^* , B ^* (both chromaticness indices) means three colorimetric values, 400 nm to 700 nm, measured by a spectrophotometer MS-2020 type colorimeter manufactured by Macbeth Co.
JIS Z8701 and Z8 from the spectral reflectance at the interval of nm
729.

First, the second module 12 corresponding to the invention of claim 1 will be described. Reference numeral 106 denotes a plurality of yarns or textile products samples such as cloth for which the same color value is obtained,
It is a color number database in which the values of a plurality of parts are measured and the values are stored.

The reference numeral 107 calculates the average colorimetric value by averaging the colorimetric values of a plurality of parts for each sample with respect to the colorimetric values of all the samples stored in the color number database 106. Is a process of calculating the difference (color difference) between the average colorimetric values for each of all the combinations of the two average colorimetric values. The color difference ΔE _{x at} this time is obtained by the following equation from the standard deviations ΔL ^* , Δa ^* , and Δb ^* of the three colorimetric values L ^* , a ^* , and b ^* .

[0014]

## EQU1 ## ΔE _x = ((ΔL ^* ) ² + (Δa ^* ) ² + (Δ
b ^* ) ² ) ^1/2 ... Formula (1)

108 is the color difference Δ obtained in step 107.
E _x is the process of selecting the two average colorimetric values is minimal. In step 109, the color difference ΔE _x between the two average colorimetric values selected in step 108 is compared with a predetermined threshold ΔE, and control is branched according to the result. That is, when the color difference is less than or equal to the threshold value ΔE, it is determined that they are the same color, and the control is moved to step 110.

In step 110, the average value corresponding to the center point of the colorimetric data of the two average colorimetric values determined to be the same color is added to the color number database as a new color number, and the two average colormetric values determined to be the same color are added. This is processing for deleting a color value. When step 110 is completed, control is returned to step 107 to form a processing loop.

On the other hand, at step 109, when the minimum color difference becomes larger than the threshold value ΔE, that is,
When two average colorimetric values having a color difference smaller than the threshold value no longer exist, the processing loop of steps 107 to 110 is exited, and the processing of the second module 12 ends.

The value stored in the color number database 106 at the time when the processing of the second module 12 is completed corresponds to the representative value of the color range which can be regarded as the same color.

Next, the first module 11 corresponding to the invention of claim 2 will be described. The first module 11 relates to a procedure for setting the above-mentioned threshold value.

Reference numeral 101 denotes a same sample colorimetric database which stores colorimetric data of a plurality of parts of the same sample. Reference numeral 102 denotes a process in which colorimetric data is read from the same sample colorimetric database, the standard deviation thereof is calculated, and the standard deviation is used as the colorimetric variation between parts. The colorimetric variation ΔS at this time is obtained from the standard deviations ΔSL ^* , ΔSa ^* , ΔSb ^* of the three colorimetric values SL ^* , Sa ^* , Sb ^* by the following formula.

[0021]

## EQU2 ## ΔS = ((ΔSL ^* ) ² + (ΔSa ^* ) ² +
(ΔSb ^* ) ² ) ^1/2 Equation (2)

Reference numeral 103 denotes a same-prescription colorimetric database that stores colorimetric data of a plurality of samples obtained by dyeing the same cloth with the same cloth / dyeing prescription at different times. 104
Is a process in which colorimetric data is read from the same fabric / dye prescription colorimetric database 103, the standard deviation thereof is calculated, and this is regarded as the dyeing reproducibility in the dyeing work. The dyeing reproducibility Δd at this time is determined by the following formula from the standard deviations ΔdL ^* , Δda ^* , Δdb ^* of the three colorimetric values dL ^* , da ^* , db ^* .

[0023]

## EQU3 ## Δd = ((ΔdL ^* ) ² + (Δda ^* ) ² +
(Δdb ^* ) ² ) ^1/2 Equation (3)

Step 105 is a process for determining the threshold value according to the color difference (ΔE) obtained by summing the colorimetric variation found in step 102 and the dyeing reproducibility found in step 104.
The color difference ΔE at this time is obtained by the following equation.

[0025]

[Equation 4] ΔE = ΔS + Δd (4)

The threshold value may be obtained based on ΔE.

Next, the third module 13 corresponding to the invention of claim 3 will be described. The third module 13 is
The following procedure is performed following step 109 of the first module 11. Step 111 is the step 107
According to the above, the color difference is calculated for all the average colorimetric values stored in the color number database 106.

In step 112, two average colorimetric values that minimize the color difference are selected in accordance with step 108,
Further, it is a process of calculating the variance of colorimetric values. Step 1
In step 13, it is determined whether or not there is a significant difference in the color difference of the average colorimetric values selected in step 112, and control is branched according to the result. The significance determination, the two average colorimetric values X _a, the difference in population means when X _b, the variance of the colorimetric value V _a, V _b, the number of colorimetric value N _a, and N _b (T) is calculated by the following equation.

[0029]

[Number 5] _{T = (X a -X b)} / (V a / N a + V b / N b) 1/2 ... formula (5)

Then, f is calculated by the following equation.

[Equation 6] c = V _a / N _a / (V _a / N _a + V _b / N _b ) ... Formula (6) 1 / f = c ² / (N _a −1) + (1-c) ² / (N _b −1) Equation (7)

Allowable value t of the difference between f calculated here and the population mean
Table 1 shows the relationship with.

[0032]

[Table 1]

This determines t, and this t is | T |
When there is a larger value, it is judged that there is no significant difference.

In the significant difference judgment by a plurality of items like the data this time, significant difference judgment is performed for all L ^* value, a ^* value and b ^* value, and when there is no significant difference in all three values, it is judged as the same color. If there is a significant difference even with 1 value, it is determined that the colors are not the same. When it is determined in step 113 that the colors are the same, the control proceeds to step 114, and when it is determined that the colors are not the same, the control proceeds to step 115.

In step 114, the same process as in step 110 is performed, and the average value corresponding to the center point of the colorimetric data of the two average colorimetric values determined as the same color is set as a new color number.
This is a process of adding to the color number database, deleting two average colorimetric values determined to be the same color, and shifting the control to step 115.

In step 115, the control is returned to step 111 to form a processing loop for determining the significant difference in steps 111 to 114, but like the threshold same color determination in steps 107 to 110, the same color determination result is used. You cannot get out of the processing loop. This is because, unlike the threshold value, the significant difference considers the variance of the data, and even if it is determined that there is a significant difference between samples of a certain color difference, it is significant if the variance is large even between samples of a larger color difference. This is because there is a case where there is no difference, and it is necessary to make a brute force determination.

However, if the control shifts because it is judged in step 113 that the colors are not the same color, the color number database has not been updated and the control is directly executed in step 111.
Even if it is returned to, the same processing is repeated. Therefore, in step 115, the color numbers determined to have a significant difference are stored, and in step 111, the color difference calculation is not performed for the combination of the stored color numbers.

Further, the processing loop of steps 111 to 114 is exited when all combinations of the color number database are judged to have a significant difference and there is no combination for calculating the color difference. Finally, the result of integrating the same color is stored in the color number database.

Next, a specific example of the present invention will be described in 3 of pure wool dyed yarn.
A description will be given based on an example of 00 color number. First, one color number is taken out from the sample, colorimetry is performed on 10 parts, and the colorimetric value is stored in the same sample colorimetric database 101.
Even if the number of colorimetric areas is too large, it does not affect accuracy, so
About 10 sites are preferable. The colorimetric values are shown in Table 2.

[0040]

[Table 2]

Further, ten samples obtained by dyeing the same material yarn with the same dyeing recipe at different times are subjected to color measurement once each and stored in the same recipe colorimetric database 103. The colorimetric values are shown in Table 3.

[0042]

[Table 3]

Next, substituting the colorimetric values of Table 2 into the equation (2) gives a colorimetric variation ΔS of 0.025, and substituting the data of Table 3 into the equation (3) gives a dyeing reproducibility Δd of 0.323. Then, by substituting these values into the equation (4), the threshold value ΔE for determining the color difference between any two color samples becomes 0.35.

Next, all the samples of 300 color numbers to be judged are subjected to color measurement for each of five parts, and the color numbers and color measurement data attached to each sample are stored. The number of colorimetric sites is 5, which is the minimum number of sites for determining the same sample to be the same color, and the minimum site for determining different color samples that are very close in the color difference diagram to be different colors. This is because both verification results using the colorimetric value of 5 are satisfied.

This color number database was read and the same color judgment (107 to 110) was made by the threshold value ΔE, and the same color judgment (111 to 115) was made by the significant difference judgment. An enlarged view is shown in FIGS. 4 and 5 by extracting a part from the aggregation result, and L ^{* of the} extracted samples A to G,
Table 4 shows the average value and standard deviation of a ^* and b ^* .

[0046]

[Table 4]

In FIGS. 4 and 5, the vertical axis represents ΔH / T _H and the horizontal axis represents ΔC / T _C. At this time, the coordinates are calculated by setting the origin at _Lo ^* = 1.0, _ao ^* = 1.5, and _bo ^* =.
-4.0, colorimetric values L ^* , a ^* of each sample,
Expressed in ^{_{b *, C o * = (}} (a o *) 2 + (b o *) 2) 1/2 C * = ((a *) 2 + (b *) 2) 1/2 T C = 0.0638C _o ^{* / (1} + 0.0131C _o
^{*) +0.638 ΔC * = C *} -C o * h o = tan -1 (b o * / a o *) If h _o is 164 ° ~345 ° t = 0.56+ | 0.2cos ( h _o +168) | If h _o is other than 164 ° ~345 ° t = 0.36+ | 0.4cos (h o +35) | f = ((C o *) 4 / ((C o *) 4 +190
_{^{0)) 1/2 T H = T}} C (tf + 1-f) ΔL * = L * -L o * ΔE * = ((a * -a o *) 2 + (b * -b o *) 2
+ (ΔL ^* ) ² ) ^1/2 ΔH ^* = ((ΔE ^* ) ² − (ΔL ^* ) ² − (ΔC ^* )
² ) ^1/2 .

First, the color difference is compared with ΔE in ascending order to perform color integration by threshold value determination. In FIG. 4, since the distance between points A and B (0.05) having the minimum color difference is smaller than ΔE (0.35), X1 (L ^* = 12.794, a ^* =) at the midpoint thereof.
1.594, b ^* =-4.052). next,
The D and E points are set to Y1 (13.337, 1.518, -3.9).
01) and combine F and Y1 points with Y2 (13.154,
1.537, -3.884), and X1 and C point are X
2 (12.834, 1.557, -4.049).

ΔE (0.3) between X2 and Y2 (0.36), between X2 and G (0.38) and between Y2 and G (0.37).
5), it becomes larger than A), B, C in the threshold judgment.
The points are integrated into X2 points, and the D, E, F points are integrated into Y2 points.

Next, color integration is performed by determining a significant difference in consideration of data variance. In Figure 5, point X2 is 5 of A, B, and C.
At the midpoint of the total 15 data (black dots) of the colorimetric data of the part,
The distribution is centered around the X2 point, and the Y2 point is also E,
A total of 15 data (black dots) of colorimetric data of 5 areas of F and G
At the middle point, the dispersion is centered on the Y2 point.

Further, the point G is also the midpoint of the colorimetric data of the five parts, and the color difference between the three points is a value larger than ΔE as described above, but from the variance, the above equations (5-7) are applied. Then, it is determined that there is a significant difference between G and X2, Y2, and there is no significant difference between X2 and Y2. Therefore, set X2 and Y2 to the midpoint Z
(12.994, 1.547, -3.966). That is, finally, the samples A, B, C, D, E,
It is determined that F has the same color and sample G does not have the same color.
Table 5 shows the respective color differences of A, B, C, D, E, and F at this time.

[0052]

[Table 5]

The color difference can be said to be the same because it is within the "allowable value of the color difference of 0.6 which can be strictly judged by a person" described in "Personal Proposal of Color Tolerance" published by Japan Color Research Institute. .

[0054]

According to the present invention, the color difference can be numerically calculated based on the colorimetric value of each sample, and the threshold value logically calculated from the colorimetric value is used as the judgment standard. The reproducibility of the same color judgment represented by the judgment of color matching / mismatching is significantly improved.

Further, since the same color can be removed from the enormous color samples that have been created randomly, and only the samples having different colors can be practically used, it is not necessary to own many samples and resipes, and management is possible. It is easier and requires less storage space. Furthermore, if the same color as the ordered color is retrieved from the past sample and dyed according to the re-sipe, it is possible to save the labor of producing a new color and reduce the cost.

[Brief description of drawings]

FIG. 1 is a processing flow chart of the present invention.

FIG. 2 is a graph of the same-color determination result in the conventional CCC Shade Sorting.

FIG. 3 is a graph of same color determination results in a conventional 555 Shade Sorting.

FIG. 4 is a graph of the same color judgment result by the threshold judgment of the present invention.

FIG. 5 is a graph of the same color judgment result by the significant difference judgment of the present invention.

Claims (3)

[Claims] 1. A plurality of textile products are color-measured at a plurality of parts for each single textile product, and the colorimetric values are stored in a color number database. A method for calculating a same color value, characterized in that the colorimetric values are integrated in a procedure. (A) For all combinations of arbitrary two average colorimetric values stored in the color number database after averaging the colorimetric values for each textile product to obtain the average colorimetric value and storing it in the color number database Calculate the color difference. (B) Two average colorimetric values with the smallest color difference are selected. (C) The color difference between the two selected average colorimetric values is compared with a predetermined threshold value to determine whether the colors are the same. (D) When it is determined that the colors are the same, a value obtained by averaging two average colorimetric values is newly stored in the color number database, and
The two selected average colorimetric values are deleted from the color number database, and the procedure returns to step (a). (E) If it is determined that the colors are not the same, the average colorimetric value stored in the color number database is used as the representative value of the same color range. 2. The threshold value according to claim 1, the deviation value of the colorimetric values of a plurality of parts of a single textile product, and the same kind of textile product dyed at different times under the same conditions. A method for calculating the same color value, which is determined according to an added value of a deviation value of the colorimetric value. 3. The procedure (e) according to claim 1, wherein when it is determined that the colors are not the same, whether the color difference has a significant difference is determined by the average and variance of the colorimetric values stored in the color number database. The same color value calculation method characterized by performing the same process as when the same color is determined if there is no significant difference.