JP2740412B2 - Color difference judgment device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for measuring the color of an object and determining whether or not a color difference from a reference color is within an allowable range.

[0002]

2. Description of the Related Art Generally, in a dye factory and a paint factory, expressing a target color on a product is an important technical requirement. Even if it is difficult to completely match the colors of these products with the target colors, it is necessary that the colors be within a range that can be recognized as the same by a human.

Conventionally, in a dyeing factory, for example, when a pass / fail judgment is made to consider that a dye evaluation product (hereinafter referred to as a sample) is the same as a target color, a technician having many years of experience has Pass / fail judgment is made based on intuition.
In recent years, using a colorimeter such as a spectrophotometer and a color difference meter,
Target color and sample color specification values (values that represent colors using numerical values)
Is obtained, and the result of the measurement is used in combination with the visual judgment of a skilled person, thereby performing a double judgment method for judging the acceptability of the sample.

[0004] However, the colorimetric values output from the colorimeter are:
Because it is not always correlated with human sensation,
Various proposals have hitherto been made to correlate with human visual judgment. For example, Japanese Patent Application Laid-Open No. 57-72027
Japanese Patent Application Laid-Open No. H11-15764 discloses a method of converting XYZ among three color attribute measurement values (hereinafter, referred to as metric three attributes) into a hue value H ^* and a saturation value C ^* via Lab to perform color determination. In Japanese Patent Application Laid-Open No. 2-74832, after converting RGB of the metric 3 attributes into Lab, it is converted into a vector having the tone angle T ゜, the tone length TR, and the hue angle H ゜ as elements, and color judgment is performed. I have.

In the above, the color difference formula (1976 CIE) L ^* a ^* b ^* recommended by the CIE (International Commission on Illumination) is used as the metric three attributes. In addition, the tolerance between the three attributes is matched. CMC (Color Measure)
(Ment Committe) color difference formula has also been devised.

[0006]

ΔE = ((ΔL ^* / l · S _l ) ² + (ΔC ^* / l _c · S _c ) ² + (ΔH / S _H ) ² ) ^0.5 This is a complete CIE1976 L ^* a ^* b ^* Since it is not a uniform color difference space, L ^* a ^* b ^* is corrected by a coefficient for expanding and compressing the difference between the three attributes so as to be visually recognized.

In the Marks & Spencer color difference equation of ICS, UK, and JP-A-63-142225 and JP-A-64-26115, metric 3 attributes are defined as density, hue difference, and sharpness difference. We propose our own color difference formula. Further, in Japanese Patent Application Laid-Open No. 3-209137, (1)
976CIE) The pass / fail judgment is made by using fuzzy inference while using the L ^* a ^* b ^* color difference.

[0008]

However, JP-A-57-72027 and JP-A-2-74832 disclose only adjustment and conversion of a metric 3 attribute.
A pass / fail judgment cannot be made mechanically using the converted value. Therefore, when a pass / fail judgment is made using these conventional techniques, an allowable color difference must be determined for each of the three metric attributes. However, in such a method, the allowable range must be reset each time the color to be determined changes, so that it is difficult to easily perform highly accurate color management. Furthermore, the method using a unique color difference formula is a global standard.
Since L ^* a ^* b ^* recommended by E is not used, it is difficult to explain the basis of the pass / fail judgment to the third party. this is,
The same applies when applying intelligent information processing technology such as fuzzy or neural network to the pass / fail judgment. As described above, a color difference equation that is not completely correlated with visual observation has not been derived yet, and at present, various kinds of pass / fail determination methods are employed.

The present invention is intended to solve such a problem. Even if a standard color system such as the CIE color difference formula is used, it is possible to judge whether or not the color difference is acceptable simply by setting a simple color difference tolerance. An apparatus is provided.

[0010]

SUMMARY OF THE INVENTION The present invention is an apparatus for determining whether or not a color difference between a predetermined reference color and a measurement color is within a predetermined allowable reference, and comprising: A colorimeter for measuring a color, a reference color storage for storing a reference color, a correction value storage for storing a correction value predetermined corresponding to the type of the object to be measured, and a measurement by the colorimeter. Correction means for correcting at least the brightness based on the values stored in the correction value storage means and the reference color storage means, and the corrected color 3 attribute measurement value corrected by the correction means in the reference color storage means. Correction color 3 compared with the stored value
A comparison means for making a pass / fail judgment based on whether or not all values of each element of the attribute measurement value are within a predetermined allowable range.
In the correction means, the correction of the lightness is performed based on the measured color and the reference.
And said correction value and the (reference color lightness value) ^-1/2 to the difference between the color
Multiply and add the result to the lightness value of the measured color
A color difference determining apparatus that calculates a corrected lightness value by using the following formula .

[0011]

The present invention corrects (CIE1976) lightness L ^* equivalent in order to cancel the color difference illusion that the surface shape of the sample gives to the eyes. Next, the total color difference of the sample is
It is decomposed into each element of the metric 3 attribute, and a judgment is made for each attribute based on a certain color difference.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings.
FIG. 1 is a configuration diagram showing the device of the present invention. In the figure, reference numeral 1 denotes an object to be measured, the color of which is measured by a colorimeter 2.

Although various types of colorimetric means are known and many of them can be used in the present invention, in this embodiment, the reflected light from the object 1 is measured by the optical sensor (21). This is converted into L ^* a ^* b ^* by a Lab conversion unit (22) using a known conversion formula and output. On the other hand, the reference color is stored in the reference color storage means 3 with metric 3 attributes such as XYZ, RGB, or L ^* a ^* b ^*, but if not directly stored in L ^* a ^* b ^* , It is preferable to provide a Lab conversion unit as in the case of the color unit 2. The setting of the reference color may be performed by a known input device 4 such as a keyboard or the like, and the output value may be stored by measuring the color of the reference sample using the same device as the colorimetric unit 2.

The value measured by the measuring means 2 needs to be corrected and then compared with a reference color.
The correction of the colorimetric value is performed by the correction unit 5.
A correction unit 5 configured to convert the colorimetric values into a total color difference vector based on the reference color (51); a brightness correction unit (52) configured to perform brightness correction on the color difference vectors;
The color difference calculating section (53) calculates each color difference of the metric 3 attribute from the corrected total color difference.

The details of each section will be described below. FIG. 2 is a plot diagram in which a colorimetric value and an example of a reference color are plotted in order to explain the function of the all-color-difference vector conversion unit (501). (CIE1976) L ^* a ^* b ^* , The reference color and the measured chromaticity of the sample are plotted. Next, the three-dimensional orthogonal coordinates are converted into cylindrical coordinates.
Thus, the circumferential direction of the cylindrical coordinates of the cylindrical coordinate, the hue value (H ^*), the radiation direction from the origin is the chroma value (C ^*). Reference color and sample, H of each L ^* a ^* b ^* value
The projection point (hue value, saturation value) on the ^* axis and C ^* axis and the L ^* value of the non-projection are adopted as the color values used in the color difference determination of the present invention. Here, the measured hue value, saturation value,
The lightness values are H ^* _S , C ^* _S , and L ^* _S , respectively, and the measured hue value, saturation value, and lightness value of the sample are respectively
Let H ^* _t , C ^* _t , L ^* _t .

Further, among H ^* , L ^* and C ^* cylindrical coordinates,
The L ^* -C ^* plane is shown in FIG. In the figure, X is the measurement point of the reference color, Y is the measurement point of the sample, the vector from Y to X is the total color difference vector, and the distance between Y and X represents the total color difference. Since the total color difference value includes the color difference given to the measured value by the unevenness of the surface shape of the sample,
It is necessary to cancel. Therefore, at least the lightness of the measured total color difference is corrected by the lightness correction unit 52, and the chromaticity of the corrected sample is obtained.

The correction is performed using the correction value m stored in the correction value storage means 6 in advance. The correction value m is stored in a correspondence table or the like provided in accordance with the type of the object to be measured, for example, a surface shape state caused by a used material, processing conditions, and the like.For example, in the case of a plain fabric using wool yarn, 3 ". Further, the setting may be made in the well-known input unit 4 similarly to the reference color storage unit 3.

The correction by the brightness correction unit 52 is performed, for example, as follows.

[0019]

## EQU2 ## ΔL ^* a = ΔL ^* × m / (L ^* _s ) ^0.5 where ΔL ^* = L ^* _t− L ^* _s

Then, the corrected brightness value (L _t ^* ) of the sample
a) is obtained by the following equation.

Equation 3] _{^{L t * a = L s *}} + ΔL * a該補Masaaki degree value using (L _t ^* a), although the correction color difference is calculated sequentially perform such processing the color difference calculating unit (5
3).

That is, C ^* of the chromaticity of the corrected sample
From the projection on the axis, the corrected chroma difference (ΔC ^* a) of the sample is obtained by the following equation.

[0022]

[Number 4] ^{ΔC * a = ΔL * a /} tan (α) However, α: Measurement corner sandwiched between the total color difference vector and the C ^* axis. tan (α) = (L ^* _t− L ^* _s ) / (C ^* _t− C ^* _S ) C ^* _t = ((a ^* _t ) ² + (b ^* _t ) ² ) ^0.5 C ^* _s = (( _{^{a * s) 2 + (b}} * s) 2) 0.5

Then, the corrected saturation value of the sample (C
^* _ta ) is obtained by the following equation.

## EQU5 ## C ^* _ta = C ^* _t + ΔC ^* a

FIG. 5 shows how to determine the hue difference (ΔH ^* ).
This will be described with reference to FIG. FIG. 5 shows the H ^* -C ^* plane among the H ^* , L ^* , and C ^* cylindrical coordinates. In the figure, S is the measurement point of the reference color, and T is the chromaticity of the sample according to the above-described method.
The distance between S and T represents the color difference. The arc 1S-2S-3S is
The reference color iso-saturation curve, arcs 1u-2u-3u, is a sample iso-saturation curve.

Next, a method of decomposing the hue component of the color difference will be described. First, the reference color (S) and the origin O of the coordinates are connected by a straight line. (SO) This becomes the hue line of the reference color. Similarly, the sample (T) and the origin O of the coordinates are connected by a straight line.
(TO) This is the hue line of the sample. An angle between these two straight lines, that is, a hue angle difference is defined as Δα.

A tangent line 1R-2R is drawn in the reference color (S) on the arc 1s-2s-3s. Sample hue line (T
−O) and the intersection of the isochromatic line (1S-2S-3S) of the reference color is P. A perpendicular line from P to the tangent line 1R-2R is the point Q, and the distance QS is the hue difference (ΔH ^* a), which is the hue component of the total color difference. The hue difference (ΔH ^* a) is obtained by the following equation.

[0027]

ΔH ^* a = C ^* _S × Δα where Δα = tan ^(-1) (b ^* _t / a ^* _t ) -tan ^(-1) (b ^* _s / a ^* _s ) C ^* _S is It is obtained by a method similar to the following Expression 8.

As described above, when the correction metric 3 attribute is obtained, a color difference allowable range is obtained based on the reference color stored in the reference color storage means 3 and the correction metric 3 attribute.
A pass / fail judgment is made in comparison with this.

The permissible color difference range is calculated by the permissible range calculating means 7. The contents will be described next. FIG. 4 shows that L ^*
-C ^{* Indicates} the total color difference allowable range ΔE _lim on the plane. ΔE
_lim is the distance between the reference color and the sample, usually 0.3
If it is within, it is visually determined to be the same color.

The projecting the ΔE lightness axis (L ^*), the Irodoridojiku (C ^*), the lightness value (L ^*), chroma (C ^*) of the color difference allowable range [Delta] L _lim, [Delta] C _lim, and Each is calculated by the following formula.

[0031]

ΔL _lim, = ΔE _lim × sin (α) where α = tan ⁽⁻¹⁾ ((L ^* _t− L ^* _s ) / (C ^* _t− C ^* _S )) C ^* _t = (( ^{_{a * t) 2 + (b}} * t) 2) 0.5 C * s = ((a * s) 2 + (b * s) 2) 0.5

[0032]

[Equation 8] _{ΔC lim, = ΔE lim × cos} (α)

FIG. 6 shows the total color difference allowable range ΔE _lim on the H ^* -C ^* plane. First, a straight line (E-
O) is subtracted. The intersection of EO with the arc 1S-2S-3S,
From the relationship with the tangent line (1R-2R) of the reference color, ΔE
hue component of _lim [Delta] H ^* _lim Request (F-G). ΔH ^*
_lim is the distance between the perpendicular G to the tangent 1R-2R and the reference color S from the intersection of the hue line (EO) of ΔE _{lim and} the _isochromatic lines 1S-2S-3S of the reference color. This calculation is performed by the following equation.

[0034]

[Equation 9] ^{_{ΔH * lim = C * S a}} × θ However, θ: O-E and the O-S is, sandwich corner. θ = ψ−φ Here, ψ: angle between b ^* axis and OE φ: angle between b ^* axis and OS φ = tan ^(-1) (b ^* _S / a ^* _S ) ψ = ^{tan (-1) ((b *} S + ΔE × cos β) / (a * S + ΔE × sin β)) where, beta: from point E, and the perpendicular E-J to the b ^* axis, E
-S included angle β = tan ^(-1) (Δa ^* / Δb ^* ) Δa ^* = a ^* _t− a ^* _s Δb ^* = b ^* _t− b ^* _s

As described above, if the allowable range and the corrected color difference are obtained for each of the three attributes of the metric, they are input to the comparing means 8 and the following condition judgment is performed. Is determined.

ΔL ^* a <L ^* _lim ΔH ^* a <H ^* _lim ΔC ^* a <C ^* _lim ΔE ^* a <E ^* _lim However, ΔE ^* a and ΔE ^* _lim are obtained by the following equations.

[0037]

ΔE ^* a = ((ΔL ^* a) ² + (ΔH ^* a) ² + (ΔC ^* a) ² ) ^0.5

[0038]

ΔE ^* _lim = ((ΔL ^* _lim ) ² + (ΔH ^* _lim ) ² + (ΔC ^* _lim ) ² ) ^0.5

The determined result is displayed on the CRT display 9
The output is performed through a known output device such as.

Next, using a specific example, the fact that the present invention closely approximates visual judgment will be described. 13 pairs of wool yarn 2
/ 60 standard color and sample are prepared, and visually judged and (CI
E1976) Measurements were made with Lab. Table 1 shows the values.
Shown in

[0041]

[Table 1]

Next, as described above, the corrected brightness difference (ΔL
^* A), correction chroma difference (ΔC ^* a), correction hue difference (ΔH ^*
a) and a corrected total color difference ΔE ^* a, and further, a lightness allowance (ΔL ^* _lim ), a saturation allowance (ΔC ^* _lim ), a hue allowance (ΔH ^* _lim ), and a total color difference allowance ΔE ^*. got _lim . Table 2 shows the results and the results of the visual judgment.

[0043]

[Table 2]

() In ΔE ^* _lim is a color difference pass / fail judgment result. From this result, it can be seen that the color difference determination method according to the present invention correlates well with visual determination.

On the other hand, in the case of the conventional color difference, it is necessary to individually set the permissible reference for each color. For example, the permissible limit value for deep blue is about 0.221, and 0.477 for beige.
And management becomes complicated.

Further, another specific example will be described. FIG. 7 plots samples having the same total color difference from the reference color in the conventional color difference measurement method.

A line OP is an isohue line of the reference color. The samples, Q, R, and S are all on the circumference (T) that is the same distance from the reference color P. That is, all of the samples have a total color difference ΔE = 0.54, and the color differences must visually appear to be the same. However, actually, the sample is Q,
It is visually determined that the color is closer to the reference color in the order of R and S.

On the other hand, using the present invention, the corrected hue difference (ΔH
^* a), corrected brightness difference (ΔL ^* a), corrected saturation difference (ΔC ^*
a) and the results of obtaining the corrected total color difference (ΔE ^* a) are shown in Table 4.
Table 3 shows the chromaticities of the samples, Q, R, and S.

[0049]

[Table 3]

[0050]

[Table 4]

[0051] From the results shown in Table 3, correction total color difference ΔE ^* a is,
The color difference from the reference color is small in the order of Q, R, and S, which confirms that the present invention correlates well with visual observation.

[0052]

According to the present invention, the measured lightness and saturation based on the surface shape of the sample can be visually judged from the measured total color difference between the reference color and the comparative color in the three-dimensional space having the three color attributes as axes. The illusion correction of the color difference is applied to all the measured color differences, and the color difference can be decomposed into each element of the three color attributes that correlate well with the visual judgment.
Based on the constant total color difference, the determination can be made based on each allowable color difference separated into each of the three color attributes, and a color difference pass / fail determination that is highly correlated with visual determination can be performed.

Further, since the judgment is made automatically even by a skilled person who does not have much experience, the pass / fail judgment of the dyed sample can be made simply and accurately.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing an apparatus of the present invention.

FIG. 2 is a plot of a reference color and a sample on a metric 3 attribute coordinate.

[3] Sample of lightness (L ^*), explanatory view showing the contents of a chroma (C ^*) Correction

FIG. 4 is an explanatory diagram showing the contents of a method of calculating a lightness allowable range (L ^* _lim ) and a saturation allowable range (C ^* _lim ) of a sample;

FIG. 5 is an explanatory diagram showing the contents of correction of a hue (H ^* ) of a sample.

FIG. 6 is an explanatory diagram showing the contents of a method for calculating a hue allowable range (H ^* _lim ) of a sample.

FIG. 7 shows a color difference correction result according to the present invention.

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-60-142221 (JP, A) JP-A-4-121626 (JP, A)

Claims (1)

(57) [Claims] An apparatus for determining whether a color difference between a predetermined reference color and a measured color is within a predetermined allowable standard, wherein the colorimetric means measures the color of the object to be measured. Reference color storage means for storing a reference color, correction value storage means for storing a correction value predetermined for the type of the measured object, and correction value storage means for storing the value measured by the colorimetric means. A correcting means for correcting at least the brightness based on the value stored in the reference color storage means, and comparing the corrected color 3 attribute measured value corrected by the correction means with the value stored in the reference color storage means. Ri Do and a comparison means that all values of the elements of the correction color 3 attribute measured value makes the acceptance judgment on whether is within a predetermined allowable range, the brightness in the correction means
Is corrected by the correction value and the (base) difference between the measured color and the reference color.
Multiplies the quasi color lightness value) ^-1/2, measured color the multiplication result
A color difference determining apparatus for calculating a corrected lightness value by adding the corrected lightness value to the lightness value .