JPH04274716A - Photoelectric chromaticity meter - Google Patents

Abstract

PURPOSE:To improve the correcting accuracy and to reduce the instrumental error by dividing the color space into a plurality of areas and performing correction for each area. CONSTITUTION:This photoelectric chromaticity meter is provided with a plurality of spectral sensors comprised of optical filters F1-F6 and photoelectric converting elements P1-P6. A plurality of reference samples in each of a plurality of areas divided in the color space are measured by the spectral sensors. A correcting coefficient is obtained so that the sum of the color difference between the measured value and the reference value of the reference sample is minimum for every area, which is in turn stored in a correcting coefficient memory means 8. Taking the correcting coefficient into consideration for tone measured value of an object to be measured, the measuring result is operated by an operating means 7.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to a photoelectric colorimeter that has a plurality of spectral sensors each consisting of an optical filter and a photoelectric conversion element, and measures the color of a sample by spectrally incoming light from an object to be measured. In particular, the present invention relates to a photoelectric colorimeter that performs color measurement with high measurement accuracy by correcting spectral sensitivity.

0002

BACKGROUND OF THE INVENTION Today, there is a need for a colorimeter that can easily and quickly measure the color value of a sample. By the way, the Commission Internationale de l'Eclairage (CIE) specifies color matching functions for the spectral sensitivity of colorimeters, but it is extremely difficult to manufacture colorimeters with spectral sensitivities that exactly match the color matching functions. This discrepancy is the main cause of absolute value errors during measurement. Furthermore, since there are variations in spectral sensitivity even between devices of the same model, inter-device errors (hereinafter referred to as instrumental errors) occur.

Conventionally, various methods have been proposed to more accurately measure the color value of a sample. For example, the known T
In the V color analyzer, a correction value is obtained for each RGB phosphor in order to remove the interference component of the trailing portion of each RGB spectral distribution.

[0004] Also, the spectral sensor outputs x1, x2, x3,
In ..., for example, in order to make the skirting amount of output x1 0, other sensor outputs x2, x3, ... are used, and the sensor output of tristimulus value type photoelectric colorimeter
In x1+x2, use kz instead of x1, and X=
It is known that the value is calculated as kz+x2.

Furthermore, there are methods in which the chromaticity region is divided into a plurality of regions, correction coefficients are determined for each region during calibration, and correction is applied to the measured values, and spectral sensitivities of the x, y, and z sensors are adjusted by other factors. It has been proposed to perform correction by multiplying the spectral sensitivities of two sensors by a constant (Japanese Patent Laid-Open No. 142239/1983).
(Japanese Patent Application Laid-open No. 2-45718).

[0006] Furthermore, there has been proposed a method for determining coefficients for a plurality of types of samples so that the sum of the color differences between the chromaticity of an image on a video printer and the chromaticity of an image on a CRT is minimized (COLOR CORRECTION). AND IN
TERPOLATION OF TELEVISION
PICTURES FOR COMPACT PRI
NTING SYSTEMS).

[0007]

[Problem to be Solved by the Invention] The above correction method is commonly used and can be expected to have a certain degree of measurement accuracy, but it is a correction to obtain an average effect over the entire color space. Since the number of calibration points is small, the correction is rough, and a sufficient effect cannot be obtained, especially when measuring only the color of a specific area. In this case, if several thousand proof colors are used, for example, desired accuracy can be expected, but this is not practical in the actual production process from the viewpoint of cost and man-hours. Furthermore, the above-mentioned conventional method does not take into consideration the reduction of the above-mentioned instrumental error.

Furthermore, the colorimeter described in JP-A-62-142239 does not correct the spectral sensitivity itself, and therefore does not provide a sufficient effect. Also, JP-A-2-
Although the colorimeter described in Publication No. 45718 corrects the spectral sensitivity itself, it is corrected in a linear color space, so it is a non-linear color space that is most commonly used in colorimeters. In the color difference space, there is a problem in that the instrumental error is not necessarily reduced in the same way in all regions.

The present invention has been made in view of the above, and
It is an object of the present invention to provide a photoelectric colorimeter that divides a color space into a plurality of regions and performs correction for each region to improve correction accuracy and reduce instrumental error.

0010

[Means for Solving the Problems] The present invention provides a photoelectric colorimeter having a plurality of spectral sensors each consisting of an optical filter and a photoelectric conversion element, in which a plurality of reference samples are detected in each region of a color space divided into a plurality of regions. Measured with the above spectroscopic sensor,
a correction coefficient storage means for storing a correction coefficient obtained by minimizing the sum of color differences between the obtained measurement value and the reference value attached to the reference sample for each region; and calculating means for calculating a measurement result by adding the correction coefficient to the measured value.

[0011] Also, representative point storage means for storing representative points representative of the chromaticity points of the reference sample for each area, and an interpolation function whose parameter is the distance between each of the representative points and the measurement point of the measurement object. The correction coefficient may be corrected using the interpolation function, and the corrected correction coefficient may be guided to the calculation means (Claim 2). In this case, it is preferable to interpolate the correction coefficient using a parameter that is inversely proportional to the square of the distance (claim 3).

0012

According to the present invention, reference values are assigned in advance to a plurality of reference samples in each region of a color space divided into a plurality of regions. Then, the photoelectric colorimeter to which correction is to be applied measures the reference sample, and a correction coefficient is determined so that the sum of the differences between the obtained measurement value and the reference value is minimized for each region, This correction coefficient is stored in the correction coefficient storage means. Thereafter, the correction coefficient is added to the measurement value obtained by measuring the measurement object, and the measurement value after this addition is taken as the measurement result.

Further, according to the invention as claimed in claim 2, a representative point is determined for each region, and even if the measurement point of the measurement object and the representative point do not match, the representative point and the measurement point are Interpolation processing is performed to correct the correction coefficient by using an interpolation function that uses the distance between points as a parameter. Therefore, the correction coefficient becomes continuous. Then, by adding the corrected correction coefficient to the measured value, a measured value as a measurement result is obtained. Furthermore, according to the third aspect of the invention, the correction coefficient is interpolated and corrected using a parameter that is inversely proportional to the square of the distance between each representative point and the measurement point.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing an embodiment of a photoelectric colorimeter according to the present invention.

[0015] F1 to F6 are filters, F1 to F3 are arranged at positions where only the reflected light from the sample 1 (measurement object) enters, or have a configuration for that purpose, and F4 to F6 irradiate the sample 1. It is arranged at a position where only light from a light source 2 such as a xenon lamp that emits light similar to natural light is incident, or with a configuration for that purpose. Above filter F
Among 1 to F6, filters F1 and F4, filters F2 and F
5. Filters F3 and F6 each have corresponding spectral sensitivities. P1 to P6 correspond to filters F1 to F6, and are photoelectric conversion elements such as photodiodes, which generate a current proportional to the light intensity of light transmitted through each filter,
They are arranged so as to face the rear surfaces of the corresponding filters F1 to F6, respectively. The combination of the filters F1 to F6 and the photoelectric conversion elements P1 to P6 form three spectral sensors that can obtain spectral sensitivity characteristics that are as close as possible to the color matching function specified by the CIE (with a small deviation). . The reason why the light itself from the light source 2 is measured is because the emitted light color of the light source affects the color judgment of the sample, and the process for correcting this influence is performed by the first calculation unit, which will be described later. 5 using a known method.

Reference numeral 3 denotes a current-voltage conversion circuit that converts each output current value from the photoelectric conversion elements P1 to P6 into voltages. 4 is an A/D conversion circuit that converts the output voltage of the current-voltage conversion circuit 3 into a digital value. 5 is a first calculation unit that performs the above processing and converts the input digital values into measured values X, Y, and Z as tristimulus values;
Z is processed based on the color space conversion formula etc. stored in the color information storage section 6. Reference numeral 7 denotes a second calculation unit that performs correction and color space conversion, which will be described later, on the measured values X, Y, and Z obtained by the first calculation unit 5, which are stored in the correction coefficient storage unit 8. Correction using correction coefficients etc. determined in advance,
Alternatively, interpolation is further performed and conversion to a required color space is performed. Reference numeral 9 denotes a display section for displaying the measured value as the measurement result obtained by the second calculation section 7.

Reference numeral 10 denotes a control section that comprehensively controls the operation of the photoelectric colorimeter, and is composed of, for example, a microcomputer. This control unit 10 includes a keyboard 11 for key-inputting various data necessary for measurement, a color space conversion system that stores color space conversion in the first and second calculation units 5 and 7, and other necessary programs, etc. Program 12 is connected. The control unit 10 also uses a reference value data of a reference sample having a reference color inputted from the keyboard 11 etc., and a program for calculating a correction coefficient, which will be described later, and an interpolation function for interpolating the correction coefficient. and a built-in ROM with programs for adding corrections to measured values.
etc. (not shown). Reference numeral 13 is driven and controlled by the control section 10 to turn on and off the power source 2. Note that instead of the program for calculating the correction coefficients, the calculated correction coefficients may be stored.

Next, the tristimulus values X, Y, Z on the colorimeter
, that is, a method of correcting the color matching functions x(λ), y(λ), and z(λ) will be described below. Note that the explanation applies to any isochromatic space, but here, as an example, L*a
This will be explained using *b* space.

Now, if the tristimulus values finally obtained for a certain colorimeter are Xc, Yc, and Zc, then Xc=X
+k11X+k12Y+k13Z...(1) Yc
=Y+k21X+k22Y+k23Z...(2)
Zc=Z+k31X+k32Y+k33Z...(
3) The correction coefficients k11 to k33 will be found.

First, a reference value for a reference sample is determined. That is, the a*b* plane is divided into M regions, for example, 8 regions I to VIII as shown in FIG.
As shown in region I of this a*b* plane, N reference samples having chromaticity points (indicated by circular marks in the figure) that are preferably evenly scattered are prepared. Similarly, for other regions, a required number of reference samples having chromaticity points evenly scattered within each region are prepared. Now, to explain region I, each reference sample in region I is measured with a master colorimeter, and (L*1R, a*1R, b*1R), ……, (L *
Obtain the values of NR, a*NR, b*NR). This value becomes a standard value, and is assigned a value to each standard sample.

Hereinafter, a procedure for determining the above-mentioned correction coefficient, which is carried out, for example, during the production process, will be explained with reference to the flowchart shown in FIG. First, in order to sequentially designate the M areas starting from area I, set i=1 (steps #1 and #2),
A procedure for determining a correction coefficient is started. Next, with any colorimeter whose color matching function is to be corrected, N
Measurements are carried out on reference samples of
(L*1,a*1,b*1),…………,(L*N,a
*N, b*N) measurement values are obtained. (Step #3).

[0022] When the measurement is completed, the following computation for calculating the correction coefficient is executed in step #4. That is,
Appropriate initial values are set for the correction coefficients k11 to k33 in the above formulas (1) to (3), and then the measured values (L*1, a*1, b*1), . . . ..., (L*N,a*
N, b*N) as (X1, Y1, Z1), ......, (XN
, YN, ZN). Next, using the above correction coefficients k11 to k33, the above (1) to (
3) Execute the correction calculation of equation (X1C, Y1C, Z1C
), ......, (XNC, YNC, ZNC) and further convert these values into (L*1C, a*1C, b*1C),...
..., (L*NC, a*NC, b*NC) and return to the original a*b* plane.

The sum of the color differences ΔE*ab between the chromaticity points of the N reference samples valued above and the chromaticity points after the above correction is expressed as:

[
0024

[Math 1]

It is determined from equation (4).

Next, other appropriate values are entered into the correction coefficients k11 to k33, and the sum of the color differences ΔE*ab is determined by the above equation (4) in the same manner as above. Hereinafter, such a procedure is repeated multiple times to calculate correction coefficients k11 to k33 when the sum of the obtained color differences ΔE*ab becomes the minimum, and these correction coefficients k11 to k33 are used as correction coefficients (k11)I to ( k
33) Store it in the correction coefficient etc. storage unit 8 as I (step #5). If i is not M (NO in step #6),
Increment i by 1, and execute the series of procedures (steps #3 to #5) executed for area I above for area II. In this way, the same process is repeated for areas III, IV, . . . , and by performing it for all M areas (YES in step #6), correction coefficients (k11)I to (k33)I, . ..., (k11
) VIII to (k33) VIII correction coefficient storage unit 8
The storage process ends (step #7).

Through a series of operations to obtain such correction coefficients, the above correction coefficients (k11)I to (k33)I, ......, (
k11) VIII to (k33) VIII are stored in advance in the correction coefficient storage section 8 within the colorimeter body as described above.

Next, correction of measured values during actual measurements using this colorimeter will be explained with reference to the flowchart of FIG. First, measure the object to be measured,
Calculate tristimulus values X, Y, Z (steps #11, #1
2). Next, the tristimulus values X, Y, and Z calculated above are calculated as L*a
Convert to a value in the isochromatic space of *b* (Step #13)
. The second calculation unit 7 determines to which region of regions I to VIII the sample belongs based on the converted values of a* and b* (step #14). Manual input is also possible when the area to which it belongs can be determined in advance. This second calculation unit 7
is the correction coefficient (k11)i~(k33) of the determined area
i is read from the correction coefficient etc. storage section 8 (step #
15), tristimulus values X, Y, Z obtained by the above measurements
Perform correction calculations according to formulas (1) to (3) above,
Calculate XC, YC, and ZC as tristimulus values after correction (
Step #16). Next, these tristimulus values XC, YC,
ZC is the value L*C, a*C, in the isochromatic space of L*a*b*
b*C (step #17), and displayed on the display unit 9 as the converted measurement result as it is or in another display form (step #18). By performing such correction in each colorimeter, the instrumental error when a required sample is measured with a plurality of colorimeters is similarly reduced in all regions of the isochromatic space.

Next, a method of performing the above correction continuously over the entire color space will be explained.

This correction calculation is performed using the correction coefficients determined for each region of the color space. This eliminates discontinuities that occur at the boundaries of regions and further reduces measurement errors for samples with chromaticity points near such boundaries.
This method interpolates correction coefficients between regions using a function of the distance between chromaticity points.

Similar to the above correction method, regions I to VII are
Correction coefficient of I (k11) I ~ (k33) I, ......, (
k11) VIII to (k33) VIII are determined and stored in the correction coefficient storage section 8 within the colorimeter body. At the same time, representative points (a*ti, b*ti) (i=1 to 8) as chromaticity points representing regions I to VIII are also stored in the correction coefficient storage section 8 in advance. This representative point is a specific point.
In addition to one chromaticity point, the positional average of reference points within one area may be used. FIG. 5 is a diagram for explaining the representative points in such a case, and the representative points are indicated by triangular marks.

FIG. 6 is a flowchart showing the procedure for interpolating the correction coefficients. In addition, in this example, the number M of regions will be described as eight.

First, the measurement object is measured (step #21), and the tristimulus values Xm, Ym, and Zm are calculated (
Step #22). Next, the calculated tristimulus values Xm, Ym
, Zm to L*ma*mb*m (step #23). Then, the chromaticity points (a*m, b*m) on the obtained a*b* plane and the representative points (a*ti, b*ti) of each of the aforementioned regions I to VIII (i=1 to 8) From equation (5), the distance di from

[0034]

[Math 2]

[0035] is determined (step #24). This distance d
i and the correction coefficient (k11) of each stored area i~
(k33) Interpolation calculation is performed using the following equation (6) using i (i=1 to 8) (step #25).

[0036]

[Math 3]

k'lm obtained by the above formula is the sample (a*m, b
*m) are the optimum correction coefficients k'11 to k'33. Using this correction coefficient, the corrected tristimulus values Xm, Ym, Z
Find m from equations (1) to (3) above (step #26
). Next, these tristimulus values Xm, Ym, Zm are L*a*
b* is converted into values L*m, a*m, b*m in the isochromatic space (step #27), and displayed on the display unit 9 as the converted measurement result (step #28).

The above interpolation method uses, for example, L*a*b
*Although the number of regions M = 8 in the space, the applied isochromatic space, the number of divided regions, and the division method are not limited to this example, and a required form may be adopted depending on the purpose etc. I can do it. Furthermore, the distance function used when interpolating the correction coefficient is not limited to the function (1/x) to the nth power (n is an integer), but can be extended to other functions that use the distance d as a parameter, for example. Note that in the function of this example, n
When =2, smoother interpolation can be performed compared to when other functions are used, and the correction effect is large.

In addition, the value assigned when measuring a reference sample with a master colorimeter may be the average value of not only one master colorimeter but multiple colorimeters, in addition to the normally assigned value, or the value determined by the CIE regulations. It may be a chromaticity point calculated from a 2° visual field, a 10° visual field, a color matching function and standard light, or the spectral reflectance of the reference sample measured by some method.

Furthermore, the representative point of each region of the color space when interpolating correction coefficients is not limited to the average value of the reference points of each region, but also the representative point of the reference sample having a chromaticity point close to the average value. It can also be the value itself. In this way, the process of determining the representative value becomes easy. Moreover, the colorimeter to which the present invention is applicable is not limited to the tristimulus value type, but can be expanded to color measuring devices having a plurality of sensors.

FIGS. 7 to 12 are diagrams for explaining measurement accuracy; FIG. 7 shows the case without correction, FIG. 8 shows the case when the conventional method has been corrected, and FIG. This shows the case where

In FIGS. 7 to 9, 100 is the reflected light from a sample such as BCRA tile RED with spectral reflectance R(λ), and 200 is the color matching function, for example, the master colorimeter's y(λ). The spectral sensitivity curve is shown. 200a shows variations in the spectral sensitivity curves of each colorimeter. In addition, in the figure, the spectral sensitivity curve 200 is the spectral sensitivity curve 200 with variations.
It is included approximately in the center of a and is not visible.

The measurement results obtained when the above sample was measured are shown in FIGS. 7(2) to 9(2). In order to see the variation in the measured values, (y(λ) −y(λ)
Define the quantity "master colorimeter") × R (λ),
The variation of each colorimeter is expressed in % when the measurement value with the master colorimeter is used as a reference (0%). Figure 7 above (
2) to FIG. 9(2), it can be seen that the variation is reduced most when using the correction method according to the present invention.

FIGS. 10 to 12 are L*a
Chromaticity distribution in *b* space, shown in Figures 7 to 9, respectively.
This corresponds to variations in the spectral sensitivity curve 200a. In FIG. 12 obtained by the correction method according to the present invention, the instrumental error is reduced compared to FIG. There is. That is, it can be seen that the instrumental error is reduced and the absolute value accuracy is also improved.

[0045]

[Effects of the Invention] As explained above, according to the present invention,
Multiple reference samples in each region of a color space divided into multiple regions are measured using the above spectroscopic sensor, and the sum of the color differences between the obtained measurement value and the reference value attached to the above reference sample is the minimum for each region. The present invention includes a correction coefficient storage means for storing the correction coefficient obtained as follows, and a calculation means for calculating the measurement result by adding the correction coefficient to the measurement value obtained by measuring the measurement target. The spectral sensitivity curve itself is corrected for each color gamut,
Compared to the conventional method that performs correction over the entire area, this method enables more accurate correction, is particularly effective when the area to be measured is limited, and reduces errors in isochromatic space. For correction purposes, it is possible to perform corrections that match visual perception.

Furthermore, since the correction coefficients are interpolated to make them continuous, correction without discontinuities can be performed over the entire color space, and instrumental differences can be further reduced. Furthermore, by making the interpolation of the correction function inversely proportional to the square of the distance between the representative point and the measurement point, smoother interpolation can be performed compared to when other functions are used.
It is possible to increase the correction effect.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of a photoelectric colorimeter according to the present invention.

[Figure 2] N with chromaticity points within the area of the a*b* plane
M regions in which reference samples are shown, e.g. regions I to V
It is a figure of the a*b* plane divided into 8 like III.

FIG. 3 is a flowchart illustrating a procedure for determining a correction coefficient, which is performed, for example, during a production process.

FIG. 4 is a flowchart illustrating correction of measured values during actual measurement.

[Figure 5] a for explaining representative points in each area
It is a diagram of the *b* plane.

FIG. 6 is a flowchart illustrating a procedure for calculating a measured value by interpolating correction coefficients.

FIG. 7 is a diagram for explaining measurement accuracy, and shows a case without correction.

FIG. 8 is a diagram for explaining measurement accuracy, and is a diagram after correction of the conventional method is applied.

FIG. 9 is a diagram for explaining measurement accuracy when correction is performed by the method according to the present invention.

10 is a diagram showing the chromaticity distribution of the measured values shown in FIG. 7 in L*a*b* space.

11 is a diagram showing a chromaticity distribution of the measured values shown in FIG. 8 in L*a*b* space.

12 is a diagram showing the chromaticity distribution of the measured values shown in FIG. 9 in L*a*b* space.

[Explanation of symbols]

1 Sample (measurement target) 2 Light source 3 Current-voltage conversion circuit 4 A/D conversion circuit 5 First calculation unit 6 Color information storage unit 7 Second calculation unit 8 Correction coefficient storage unit 9 Display unit 10 Control unit 11 Keyboard 12 Color space conversion system, etc. program 13 Illumination circuits F1 to F6 Filters P1 to P6 Photoelectric conversion elements

Claims (3)

[Claims] Claim 1: A photoelectric colorimeter having a plurality of spectral sensors consisting of an optical filter and a photoelectric conversion element, in which a plurality of reference samples in each region of a color space divided into a plurality of regions are measured with the spectral sensor, correction coefficient storage means for storing a correction coefficient obtained by minimizing the sum of color differences between the measured value and the reference value of the reference sample for each region;
A photoelectric colorimeter comprising a calculation means for calculating a measurement result by adding the correction coefficient to a measurement value obtained by measuring a measurement object. 2. The photoelectric colorimeter according to claim 1, comprising:
representative point storage means for storing representative points representing the chromaticity points of the reference sample for each region; and function storage means for storing an interpolation function whose parameter is the distance between each of the representative points and the measurement point of the measurement object. A photoelectric colorimeter comprising the following: the obtained correction coefficient is corrected using the interpolation function, and the corrected correction coefficient is guided to the calculation means. 3. The photoelectric coloring device according to claim 2, wherein the interpolation function is expressed by (1/d)2, where d is a distance between the representative point and a measurement point of the measurement target. Total.