Color measurement device, data processing device, measurement correction method, and program

Abstract

The invention relates to a colorimetric apparatus, a data processing apparatus, a measurement correction method, and a program. A color measurement device is provided with: illumination units (11, 12) that illuminate an object to be measured (100); a two-dimensional photoelectric conversion unit (133) that disperses, for each wavelength, light from a plurality of positions on the surface of the measurement object illuminated by the illumination unit, receives the light, and converts the light into an electrical signal; a calculation unit (14) that calculates a regression illumination correction coefficient for correcting an error of each pixel caused by re-illumination in the color measurement device (1), on the basis of the light reception results of a plurality of samples, the brightness of which is uniform in each part of the measurement surface and the brightness of which is different in the measurement surface, by the photoelectric conversion unit; and a correction unit (14) that corrects the measurement value of the measurement object using the regression illumination correction coefficient calculated by the calculation unit.

Description

TECHNICAL FIELD

[0001] The present application relates to a color measurement device, a data processing device, a measurement correction method, and a program suitable for measuring color, reflectance of a measurement object having a two-dimensional configuration such as a texture, and the like. BACKGROUND

[0002] As a color measurement device capable of easily measuring color, spectral reflectance of a sample having a two-dimensional configuration such as a texture, a two-dimensional color measurement device is known. The two-dimensional color measurement device is provided with an illumination unit that illuminates a measurement object, and a two-dimensional photoelectric conversion unit that spectrally splits reflected light from a plurality of positions on the surface of the measurement object illuminated by the illumination unit and converts it into an electric signal for each wavelength.

[0003] A user is able to obtain a spatial distribution of luminance, a chromaticity distribution for each wavelength by spectrally splitting and measuring the reflected light from a plurality of positions on the surface of the measurement object using the color measurement device.

[0004] In this way, in the case of using a two-dimensional color measurement device, the main purpose is to acquire a distribution of each wavelength in space. Therefore, as a color measurement device, it is possible to cite invariance in measurement values in each part of the measurement surface, in other words, in-plane uniformity, as an important performance.

[0005] That is, in the case of measuring a measurement object whose luminance is uniform in each part of the measurement surface and whose distribution of measurement values for each wavelength is considered to be substantially uniform, ideally, it is required to obtain the same measurement result at each pixel of the two-dimensional photoelectric conversion unit. However, in an actual color measurement device, it is difficult to obtain a completely uniform result only by efforts on hardware due to the following factors that hinder in-plane uniformity.

[0006] [Illumination unevenness]

[0007] In the case where the intensity of illumination of the measurement unit differs depending on the position due to unevenness in the intensity of the light source, orientation characteristics, unevenness in the reflectance / transmittance of optical components that constitute the illumination system, and the like, illumination unevenness occurs.

[0008] [Light receiving unevenness]

[0009] In the case where the light receiving sensitivity differs depending on the position due to unevenness in the reflectance / transmittance of optical components that constitute the light receiving system, unevenness in the diffraction efficiency of a light splitting element such as a diffraction grating in the case of a spectral color measurement device, and the like, light receiving unevenness occurs.

[0010] [Sensor sensitivity unevenness]

[0011] The light receiving sensitivity of a sensor used in the photoelectric conversion unit can cause unevenness in the sensitivity of each pixel (PRNU).

[0012] When these in-plane uniformity defects caused by the structural components of the measurement device occur, even if the distribution of measured values ​​for each wavelength across the measurement surface of the sample is considered roughly uniform, non-uniformity will still occur in the measurement results. Therefore, such non-uniformity should be minimized as much as possible. While hardware-level countermeasures are certainly feasible, numerical in-plane non-uniformity correction is often performed to address errors that cannot be completely eliminated by hardware alone.

[0013] For example, Patent Document 1 discloses a method for correcting measured values ​​using a known sample whose distribution of absorptivity for each wavelength in each portion of a measurement surface is considered to be substantially uniform, particularly in order to eliminate uneven illumination intensity.

[0014] Patent Document 1: U.S. Patent No. 9,784,614

[0015] However, conventional techniques have the problem of poor calibration accuracy for objects with brightness outside the calibration point. For example, conventional techniques attempt to eliminate poor uniformity by using a single sample whose distribution of absorbance at each wavelength in each part of the measurement surface is considered to be roughly uniform. However, when actually measuring samples with different brightness using a two-dimensional colorimetric device, it becomes Figure 8 The results of (a) and (b).

[0016] Figure 8 (a) shows the relative spatial intensity distribution before correction, and (b) shows the relative spatial intensity distribution after unevenness correction based on conventional technology. Specifically, based on the two-dimensional spectral reflectance when the calibration white plate and the non-color reference sample were measured in the SCI method using an integrating sphere (d:8), the ratio of the color value L* when cut at a cross section passing through the center of the measurement diameter is shown. Among them, in each sample, the standardization is to take the measurement diameter center as 100. In the chart, S1 represents the calibration white plate, and S2 to S5 represent non-color reference samples of different brightness.

[0017] like Figure 8 As shown in (a), the relative spatial intensity distribution varies significantly between samples with different brightness. Therefore, as shown in (b) of the same figure, if the correction coefficient is calculated using a moving average of 11×11 pixels on a white board and correction is performed on each block, there will be some improvement on the white board, but there will be almost no improvement in the low-brightness baseline sample S5, especially as shown by the bold line.

[0018] That is, the conventional single-point correction method can ensure in-plane uniformity in brightness near the sample (calibration point) used for correction coefficient calculation, but cannot ensure uniformity in samples with different brightness. SUMMARY

[0019] An object of the present application is to provide a color measurement device, a data processing device, a measurement correction method, and a program that can suppress errors in measurement values in each part of a measurement surface of a measurement object and achieve an improvement in accuracy of measurement results.

[0020] The above object is achieved by the following method.

[0021] (1) A color measurement device, comprising:

[0022] an illumination section that illuminates a measurement object;

[0023] a two-dimensional photoelectric conversion section that spectrally separates light from a plurality of positions on a surface of the measurement object illuminated by the illumination section and receives and converts the light into an electrical signal for each wavelength;

[0024] a calculation section that calculates a regression-illumination correction coefficient for correcting errors of each pixel caused by re-illumination within the color measurement device, based on light-receiving results of a plurality of samples that are uniform in luminance in each part of the measurement surface and different in luminance in the measurement surface, by the photoelectric conversion section; and

[0025] a correction section that corrects measurement values of the measurement object using the regression-illumination correction coefficient calculated by the calculation section.

[0026] (2) The color measurement device according to the preceding item 1, wherein

[0027] the calculation section calculates a non-uniformity correction coefficient for correcting non-uniformity of each pixel caused by structural components of the color measurement device, based on the light-receiving results of the samples by the photoelectric conversion section,

[0028] the correction section corrects the measurement values of the measurement object using the non-uniformity correction coefficient calculated by the calculation section.

[0029] (3) The color measurement device according to the preceding item 1, wherein

[0030] the calculation section calculates the regression-illumination correction coefficient for each wavelength at which spectrometry is performed.

[0031] (4) The color measurement device according to the preceding item 2, wherein

[0032] the calculation section calculates the non-uniformity correction coefficient for each wavelength at which spectrometry is performed.

[0033] (5) The color measurement device according to the preceding item 1, wherein

[0034] the calculation section calculates the regression-illumination correction coefficient for each tristimulus value.

[0035] (6) The color measurement device according to the preceding 2, wherein

[0036] The calculation unit calculates the unevenness correction coefficient for each tristimulus value.

[0037] (7) The color measurement device according to the preceding 1, wherein

[0038] The sample is a sample in which the distribution of the measured values for each wavelength is regarded as substantially uniform.

[0039] (8) The color measurement device according to the preceding 1, wherein

[0040] The calculation unit calculates, as the regression-illumination correction coefficient, a second-order coefficient a2_2(x, y, λ), a first-order coefficient a2_1(x, y, λ), and a zero-order coefficient a2_0(x, y, λ) when a quadratic function of the deviation from the reference value with the abscissa set as the spectral reflectance and the ordinate set as the spectral reflectance is approximated by Formula 1 below,

[0041] The correction unit corrects the measured values by Formula 2 below using the quadratic function of the calculated respective regression-illumination correction coefficients,

[0042] [Formula 1]

[0043] a2_2(x, y, λ) = ((Y1 - Y2) * (X1 - X3) - (Y1 - Y3) * (X1 - X2)) / ((X1 - X2) * (X1 - X3) * (X2 - X3))

[0044] a2_1(x, y, λ) = (Y1 - Y2) / (X1 - X2) - (a2_2) * (X1 + X2)

[0045] a2_0(x, y, λ) = Y1 - (a2_2) * X1 * X1 - (a2_1) * X1

[0046] where x and y represent the xy coordinates of the measurement surface, λ represents the wavelength, and, regarding the three samples, when the target measured values are set as Rc_1(λ), Rc_2(λ), and Rc_3(λ), respectively, the measured values in the color measurement device that is the correction target are set as Rs_1(λ), Rs_2(λ), and Rs_3(λ), respectively, the regression-illumination error is set as ΔRn = Rs_n(x, y, λ) - Rc_n(x, y, λ), and n = 1, 2, 3,

[0047] (X1, Y1) = (Rs_1(x, y, λ), ΔR1(x, y, λ))

[0048] (X2, Y2) = (Rs_2(x, y, λ), ΔR2(x, y, λ)) (X3, Y3) = (Rs_3(x, y, λ), ΔR3(x, y, λ))

[0049] (X3,Y3) = (Rs_3(x,y,λ), ΔR3(x,y,λ))

[0050] [Equation 2]

[0051] [Equation 1]

[0052] Ref rerefcorr(x,y,λ) = Ref0(x,y,λ) - (a2_2(x,y,λ) x Ref0(x,y,λ))2+ a2_1(x,y,λ) x Ref0(x,y,λ) + a2_0(x,y,λ))

[0053] where Ref0(x,y,λ) is a measurement value before correction, and Ref rerefcorr (x,y,λ) is a measurement value after correction.

[0054] (9) The color measurement device according to the preceding item 2, wherein

[0055] The above calculation unit calculates a mura correction coefficient Mura(x,y,λ) by the following equation when a target measurement value is set to Count target (x,y,λ) and a measurement value of the second sample is set to Count(x,y,λ),

[0056]

[0057] The above correction unit corrects the measurement value Count(x,y,λ) to a measurement value after correction Count corr (x,y,λ) by the following equation,

[0058] [Equation 3]

[0059] Conut corr (x,y,λ) = Mura(x,y,λ) x Count(x,y,λ).

[0060] (10) The color measurement device according to the preceding item 1, wherein

[0061] The above sample includes a patch of gray or black.

[0062] (11) The color measurement device according to the preceding item 2, wherein

[0063] The above sample includes a white plate.

[0064] (12) The color measurement device according to the preceding item 1, wherein

[0065] The calculation unit calculates a weighted average of the neighboring pixels and sets the weighted average as the final correction coefficient after calculating the regression illumination correction coefficient for each pixel.

[0066] (13) The color measurement device according to the preceding 1, wherein

[0067] The calculation unit approximates the regression reflection correction coefficients of the respective pixels to a polynomial when there is continuity in the regression reflection correction coefficients, calculates the coefficients of only one pixel, calculates the coefficients of all the pixels from the position coordinates of the measurement surface, and sets the coefficients of all the pixels as the final correction coefficients.

[0068] (14) The color measurement device according to the preceding 2, wherein

[0069] The calculation unit calculates a weighted average of the neighboring pixels and sets the weighted average as the final correction coefficient after calculating the uneven correction coefficient for each pixel.

[0070] (15) The color measurement device according to the preceding 2, wherein

[0071] The calculation unit calculates the correction coefficients of all the pixels from the position coordinate relationship of the known in-plane unevenness caused by the structural members of the measurement device after calculating the uneven correction coefficient of only one pixel, and sets the correction coefficients of all the pixels as the final correction coefficients.

[0072] (16) The color measurement device according to any one of the preceding 1 to 15, wherein

[0073] The light from the plurality of positions on the surface of the measurement object is reflected light from the plurality of positions.

[0074] (17) The color measurement device according to any one of the preceding 1 to 15, wherein

[0075] As the illumination unit, an integrating sphere that diffusely reflects light from a light source on an inner surface is used.

[0076] (18) A data processing device, comprising:

[0077] The receiving unit receives, from a color measurement device that includes an illumination unit and a two-dimensional photoelectric conversion unit, a light-receiving result of the photoelectric conversion unit for a plurality of samples that are different in luminance in a measurement surface, the illumination unit illuminating a measurement object, and the photoelectric conversion unit spectrally splitting light from a plurality of positions on a surface of the measurement object illuminated by the illumination unit by wavelength and converting the light into an electric signal;

[0078] a calculation unit that calculates a regression-illumination correction coefficient for correcting an error of each pixel caused by re-illumination within the color measurement device, based on the light-reception result received by the reception unit; and

[0079] a correction unit that corrects a measurement value of the measurement object using the regression-illumination correction coefficient calculated by the calculation unit.

[0080] (19) A measurement correction method, comprising:

[0081] an illumination step of illuminating, by an illumination unit, a plurality of samples that are uniform in luminance in each part of a measurement surface and different in luminance in the measurement surface;

[0082] a conversion step of spectrally splitting light from a plurality of positions on a surface of the sample illuminated by the illumination unit by each wavelength, and receiving and converting into an electric signal by a two-dimensional photoelectric conversion unit;

[0083] a calculation step of calculating a regression-illumination correction coefficient for correcting an error of each pixel caused by re-illumination within the color measurement device, based on a light-reception result of the sample by the photoelectric conversion unit; and

[0084] a correction step of correcting a measurement value of the measurement object using the regression-illumination correction coefficient calculated by the calculation step.

[0085] (20) A program for causing a computer to execute the following steps:

[0086] a reception step of receiving, from a color measurement device provided with an illumination unit and a two-dimensional photoelectric conversion unit, a light-reception result of a plurality of samples that are uniform in luminance in each part of a measurement surface and different in luminance in the measurement surface by the photoelectric conversion unit, the illumination unit illuminating a measurement object, the photoelectric conversion unit spectrally splitting light from a plurality of positions on a surface of the measurement object illuminated by the illumination unit by each wavelength and converting into an electric signal;

[0087] a calculation step of calculating a regression-illumination correction coefficient for correcting an error of each pixel caused by re-illumination within the color measurement device, based on the light-reception result received by the reception step; and

[0088] a correction step of correcting a measurement value of the measurement object using the regression-illumination correction coefficient calculated by the calculation step.

[0089] According to the color measurement device and the measurement correction method of the present application, light from a plurality of positions of a measurement surface illuminated by an illumination unit is spectrally split by wavelength and received by a two-dimensional photoelectric conversion unit and converted into an electric signal. Based on the light receiving results of the photoelectric conversion unit for a plurality of samples in which the luminance is uniform in each part of the measurement surface and the luminance differs in the measurement surface, a regression illumination correction coefficient for correcting the error of each pixel caused by re-illumination within the measurement device is calculated. The measurement value of the measurement object is corrected using the calculated regression illumination correction coefficient.

[0090] Thus, even if the measurement object is a luminance that cannot be dealt with by unevenness correction of unevenness possessed by the structural member of the color measurement device for eliminating unevenness of illumination intensity as in the past, the error of the measurement value in each part of the measurement surface can be suppressed to achieve improvement in the accuracy of the measurement result.

[0091] The data processing device of the present application receives light receiving results of the photoelectric conversion unit for a plurality of samples in which the luminance is uniform in each part of the measurement surface and the luminance differs in the measurement surface. The data processing device can perform the following processing: based on the light receiving results, calculate a regression illumination correction coefficient for correcting the error of each pixel caused by re-illumination within the measurement device, and correct the measurement value of the measurement object using the regression illumination correction coefficient.

[0092] According to the program of the present application, the computer can be caused to perform processing of receiving light receiving results of the photoelectric conversion unit for a plurality of samples in which the luminance is uniform in each part of the measurement surface and the luminance differs in the measurement surface. Also, according to the above program, the computer can be caused to perform the following processing: based on the light receiving results, calculate a regression illumination correction coefficient for correcting the error of each pixel caused by re-illumination within the measurement device, and correct the measurement value of the measurement object using the regression illumination correction coefficient. BRIEF DESCRIPTION OF DRAWINGS

[0093] Figure 1 FIG. 1 is a diagram schematically showing the structure of a color measurement device according to one embodiment of the present application.

[0094] Figure 2 FIG. 2 is a schematic structural diagram of a light receiving unit.

[0095] Figure 3 FIG. 3 is a structural diagram of the case where the calculation of the correction coefficient is performed by a data processing device.

[0096] Figure 4 FIG. 4 is a flowchart showing the main measurement flow performed by the color measurement device.

[0097] Figure 5 FIG. 5 is a graph for explaining the influence of regression illumination characteristics.

[0098] Figure 6This is a graph showing the effects of the present invention for a plurality of samples having different brightness.

[0099] Figure 7 This is another graph showing the effects of the present invention for a plurality of samples having different brightness.

[0100] Figure 8 This is a diagram for explaining problems in conventional technology. DETAILED DESCRIPTION

[0101] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0102] Figure 1 This figure schematically illustrates the configuration of a colorimetric device 1 comprising a spectrophotometer, according to one embodiment of the present invention. In this figure, colorimetric device 1 measures the spectral reflectance characteristics of an object 100 having a two-dimensional structure such as texture. Colorimetric device 1 includes an integrating sphere 11, a light source 12, a light receiving unit 13, a computing unit 14, a control unit 15, a recording unit 16, a display unit 17, and an operation panel 18.

[0103] Integrating sphere 11 is a hollow sphere whose inner wall 111 is coated with a highly diffuse and highly reflective white diffuse reflective coating, such as magnesium oxide or barium sulfate. Integrating sphere 11 is configured to generate diffusely reflected light by causing light emitted from light source 12 to reflect multiple times on inner wall 111. Light source 12 is, for example, a xenon flash lamp.

[0104] Integrating sphere 11 also has a measuring opening 112 formed at its lower end, facing object 100. Furthermore, integrating sphere 11 has a light-receiving opening 114, located opposite measuring opening 112 and extending through the sphere at an angle of 8° relative to a normal 113 to the opening surface of measuring opening 112. Furthermore, light-shielding walls 115, 115 are formed within integrating sphere 11 to prevent light emitted from light source 12 from directly striking measuring opening 111 and light-receiving opening 114. Furthermore, a portion of inner wall 111 of integrating sphere 11 serves as a reference area 116 for measuring reference light.

[0105] like Figure 2 As shown, the light receiving unit 13 includes a spectroscopic unit 131 , an imaging lens 132 , and an area sensor 133 which is a two-dimensional imaging element formed of a CCD sensor or the like.

[0106] The spectroscopic unit 131 separates the light received via the light receiving opening 114 into respective wavelengths. The imaging lens 132 forms an image of the light of each wavelength separated by the spectroscopic unit 131 on the area sensor 133 .

[0107] The area sensor 133 corresponds to a photoelectric conversion unit, such as Figure 2The area sensor 133 has a plurality of pixels 134 arranged vertically and horizontally. Figure 2 The x-direction (x-direction) means the horizontal direction of the physical space. Each horizontal pixel 134 corresponds to the horizontal area of ​​the object to be measured. On the other hand, the vertical direction ( Figure 2 In other words, each pixel 134 in the horizontal pixel column corresponds to a plurality of regions in the one-dimensional direction of the object 100 to be measured, and the light emitted from each region and decomposed into wavelengths is received by each pixel 134 in the vertical pixel column. Therefore, in order to perform spectroscopic measurement on each region in the two-dimensional direction (plane) of the object 100 to be measured, it is necessary to move the object 100 along the horizontal direction. Figure 2 Alternatively, the colorimetric device may be moved in the y direction instead of moving the object 100. Figure 2 Alternatively, the object 100 to be measured and the colorimetric device 1 may be moved in the y direction, or the object 100 to be measured and the colorimetric device 1 may be moved with a speed difference.

[0108] Furthermore, as described above, a technique for dividing the plane of the object 100 into regions of sizes corresponding to the pixels 134 of the area sensor 133 and splitting light from each region to receive light at each pixel 134 of the area sensor 133 is known as, for example, a hyperspectral camera.

[0109] The electrical signals, ie, measurement data, output from each pixel 134 of the area sensor 133 are converted into digital signals through a current-voltage (IV) conversion circuit and an analog-to-digital (AD) conversion circuit (not shown) as needed, and are sent to the calculation unit 14 .

[0110] The calculation unit 14 uses the received measurement data to calculate the color and reflectance for each of the multiple regions on the measurement object 100. Furthermore, in this embodiment, the calculation unit 14 calculates an unevenness correction coefficient and a regression illumination correction coefficient during measurement. Furthermore, the calculation unit 14 uses the calculated unevenness correction coefficient and regression illumination correction coefficient to correct the measurement data and output the corrected values. The unevenness correction coefficient and regression illumination correction coefficient will be described later.

[0111] The control unit 15 comprehensively controls the entire colorimetric device 1 and includes a CPU, a RAM, etc. In the present embodiment, the calculation unit 14 also constitutes a part of the function of the control unit 15 .

[0112] The recording unit 16 is a memory that records the calculated unevenness correction coefficient and regression illumination correction coefficient, the measured value (output value) corrected using these correction coefficients, the measured value (raw data) before correction, and the like.

[0113] The display unit 17 displays the calculation results of the calculation unit 6 , other data, messages, and the like.

[0114] The operation panel unit 18 is operated by a user when using the colorimetric device 1 .

[0115] The calculation unit 6 may also be arranged in the colorimetric device 1 as in the present embodiment. Figure 3 As shown, the calculation unit 6 can also be composed of a personal computer (equivalent to a data processing device, hereinafter referred to as PC) 2 connected to the colorimetric device 1. In this case, the measurement value output from the area sensor 133 and processed into a digital signal can be sent from the transceiver 19 of the colorimetric device 1 via the network to the transceiver 21 of the PC 2, and then input into the PC 2. Furthermore, the calculation results and the like can be recorded in the recording unit within the PC 2 or displayed on the display unit of the PC 2. With this configuration, measurements can be performed even if the PC 2 is located far from the measurement location.

[0116] Next, use Figure 1 A method for measuring a measurement value (eg, reflectance) of an object 100 using the colorimetric apparatus 1 shown will be described.

[0117] Figure 4 This is a flowchart showing the main measurement process performed by the colorimetric device 1. In this embodiment, the following processes are performed in order: A / D count value acquisition process (#1), light intensity correction process (#2), unevenness correction process (#3), level correction process (#4), regression illumination correction process (#5), and reflectance output process (#6).

[0118] The A / D count value acquisition process (#1) acquires data received by each pixel 134 of the area sensor 133, converts it into an electrical signal, and then converts it into a digital signal through the AD conversion circuit. The light intensity correction process (#2) corrects for fluctuations in the light source 12.

[0119] As will be described later, the unevenness correction process (#3) is a process for correcting unevenness of each pixel caused by components of the colorimetric device 1 and is performed using an unevenness correction coefficient.

[0120] The level correction process (#4) is a process for correcting the level of the measured value, and also includes zero point correction.

[0121] As described later, the regression illumination correction process (#5) is a process for correcting the error of each pixel 134 of the area sensor 133 caused by the re-illumination of light from the inner surface 111 of the integrating sphere 11 of the colorimetric device 1, and is performed using a regression illumination correction coefficient.

[0122] The reflectance output process (#6) is a process of outputting the measured value after each process is performed to the display unit 17, the recording unit 16, and the like.

[0123] [unevenness correction processing]

[0124] Next, the unevenness correction processing (#3) is explained. This unevenness correction processing is correction processing performed to eliminate measurement errors of each part in the measurement surface of the measurement object 100. As explained in the section of the background art, measurement errors are generated due to unevenness of the illumination, unevenness of the light reception, unevenness of the sensitivity of the area sensor 5, and unevenness of the structural components of the color measurement device 1.

[0125] First, the calculation of the unevenness correction coefficient used for the correction is explained.

[0126] A sample (first sample) in which the brightness is uniform in each part of the measurement surface and the distribution of the measurement value of each wavelength is considered to be substantially uniform is measured. As such a first sample, a white plate whose surface has no concave-convex and is flat can be cited.

[0127] The first sample (white plate) is measured, and the measurement value before the correction is set as Count(x, y, λ). Here, (x, y) represents the xy coordinates on the measurement surface, and λ represents the wavelength (nm). The xy coordinates on the measurement surface also correspond to the coordinates of the pixels 134 of the area sensor 133.

[0128] The target value of the correction is set as Count target (λ). For Count target (λ), if the target value such as the measurement value in the color measurement device that has been set to be the reference is used, it is also possible. Or, it is also possible to set by measuring the average value or the like of the measurement object region such as the diameter.

[0129] If the average value of 11 x 11 pixels at the center of the measurement diameter of the measurement object 100 is set as the target value, Count target (λ) is as follows.

[0130] [mathematical expression 4]

[0131]

[0132] The unevenness correction coefficient Mura(x, y, λ) can be calculated by division so as to match the measurement value Count(x, y, λ) of each pixel 134 of the area sensor 133 and each wavelength with respect to the second sample with the above target value as follows.

[0133] [mathematical expression 5]

[0134]

[0135] On the other hand, in the case of the above-described operation method, if dust, foreign matter, or an internal defect is attached to the first sample (white plate) at the time of measurement for calculation of the unevenness correction coefficient, there is a concern that the correction coefficient is abnormal only in that pixel.

[0136] On the other hand, unevenness caused by, in particular, unevenness of illumination and light reception is generally not a change that occurs sharply in a specific pixel, but has a slow tendency. Therefore, from the viewpoint of improving the robustness of the correction coefficient, a process of reducing the influence caused by measurement error on the correction coefficient calculated as described above can also be added. As one of the methods, a method of calculating a moving average value from a plurality of pixels adjacent to the subject pixel can be cited. Below, as an example, a case where a moving average value is calculated from ±5 pixels adjacent to the subject pixel is a correction coefficient Mura'(x, y, λ).

[0137] [mathematical expression 6]

[0138]

[0139] Using the unevenness correction coefficient Mura(x, y, λ) of each pixel (x, y) thus calculated, the error caused by unevenness of the structural member of the color measurement device 1 is corrected (unevenness correction) with respect to the measured object 100 as a measurement target. The measured value Count(x, y, λ) after correction corr (x, y, λ) is corrected, for example, as follows.

[0140] [mathematical expression 7]

[0141] Conut corr (x, y, λ) = Mura(x, y, λ) * Count(x, y, λ)

[0142] Further, as the unevenness correction coefficient, the moving average Mura'(x, y, λ) described above can also be used instead of Mura(x, y, λ).

[0143] In addition, in the case where the unevenness in the measurement surface caused by unevenness of the structural member of the color measurement device 1 is known, it can also be performed as follows. That is, the unevenness correction coefficient is calculated only for one pixel. Thereafter, the correction coefficient of all pixels can also be calculated in accordance with the positional coordinate relationship of the known in-plane unevenness caused by the structural member of the color measurement device, and set as the final correction coefficient. In this case, there is an advantage that it is not necessary to save the correction coefficient for each pixel, but only one pixel is necessary.

[0144] [Regression Illumination Correction Process]

[0145] Next, the regression illumination correction process will be described.

[0146] As already described, the unevenness correction process alone cannot eliminate the error, particularly in a sample having a brightness different from that of the first sample.

[0147] The inventors believe that the cause of the error unique to each brightness is pixel dependency in the regression illumination characteristic unique to the colorimetric device 1 .

[0148] That is, in an illumination system using integrating sphere 11 with a d:8° geometry, light radiated from the surface of object 100 being illuminated is repeatedly diffusely reflected by inner surface 111 of integrating sphere 11 , thereby generating recurrent illumination that illuminates the surface of object 100 again.

[0149] In the d:8° geometric illumination and light receiving system, the light beam emitted from the light source 12 is repeatedly diffusely reflected by the inner surface 111 of the integrating sphere 11 to become diffusely reflected illumination light, which illuminates the measurement surface of the measurement object 100 and the reference area 116 on the inner surface of the integrating sphere 11.

[0150] The amount and influence of the regressive illumination depend on the reflectivity and aperture ratio of the integrating sphere 11, the optical path from the measurement object 100 (the optical path of the sample ( Figure 1 P1)) and the optical path from the reference area 116 (reference optical path ( Figure 1 Therefore, even if the integrating sphere type colorimetric device 1 is the same, it may be different depending on the model. In particular, since the zero point correction and white point correction are performed to calibrate the values ​​of the zero point and the white point, it is known that the influence of the regression illumination characteristic generally has a quadratic function characteristic that is significantly expressed at the intermediate color between the zero point and white (see Figure 5 ).exist Figure 5 The horizontal axis represents reflectivity, and the vertical axis represents reflectivity error. The shape of the quadratic function changes depending on the position of the opening 114.

[0151] Therefore, each colorimetric device 1 measures multiple samples (second samples). Unlike the first sample, the brightness of the measurement surface of these samples is uniform across all portions of the measurement surface, and the distribution of measured values ​​at each wavelength is considered to be approximately uniform. The regression illumination characteristic inherent to the colorimetric device 1, or the coefficients of an approximate quadratic function, are then calculated as regression illumination correction coefficients. This quadratic function is then subtracted from the measured values ​​when measuring the object 100, thereby reducing the influence of the regression illumination characteristic. Achromatic samples such as white, gray, and black can be used as the second sample.

[0152] The regression illumination is, as described above, a phenomenon in which the reflected light from the measurement object 100 is repeatedly diffusely reflected within the integrating sphere 11, and then illuminates the measurement object 100 again. Therefore, particularly in the case of measuring a color measurement device having a relatively large diameter, the influence thereof differs depending on the coordinates (pixels) of the measurement surface. In addition, the regression illumination characteristic, as described above, depends on the relationship of the sample light path PI and the reference light path P2. If the sample system and the reference system are completely the same in the regression illumination characteristic, both cancel out in the process of the calculation of the reflectance or the like as the measurement object, and there is no influence as the output of the color measurement device 1.

[0153] Next, the calculation of the regression illumination correction coefficient will be described.

[0154] The regression illumination correction using three samples, sample 2-1 (white plate), sample 2-2 (gray patch), and sample 2-3 (black patch) as the second samples will be described as an example. The target reflectance of the measurement reflectance or the like in the color measurement device to be a reference is set to Rc_1(λ), Rc_2(λ), and Rc_3(λ), respectively. In addition, the measurement reflectance of each of the samples 2-1, 2-2, and 2-3 measured by the color measurement device 1 to be corrected is set to Rs_1(λ), Rs_2(λ), and Rs_3(λ), respectively.

[0155] As described above, it is generally known that the difference in the regression illumination characteristic in two different color measurement devices is a quadratic function error, and for the quadratic function error, the error is 0 at the common zero point correction point and the white correction point in the case where the horizontal axis is the reflectance and the vertical axis is the reflectance error. In addition, the characteristic is a device-specific characteristic. Therefore, the correction can be performed by measuring the samples 2-1, 2-2, and 2-3 of which the target reflectance is known by using the color measurement device 1 to be corrected, and estimating the amount of error (quadratic function) caused by the regression illumination characteristic, and subtracting the amount. Specifically, the regression illumination error is set to:

[0156] ΔRn = Rs_n(x, y, λ) - Rc_n(x, y, λ) (n = 1, 2, 3)

[0157] And three points are defined as follows.

[0158] (X1, Y1) = (Rs_1(x, y, λ), ΔR1(x, y, λ))

[0159] (X2, Y2) = (Rs_2(x, y, λ), ΔR2(x, y, λ))

[0160] (X3, Y3) = (Rs_3(x, y, λ), ΔR3(x, y, λ))

[0161] For each wavelength, each spatial pixel, a quadratic function passing through the above three points is estimated. When the second, first, and zero order coefficients thereof are set as a2_2(x, y, λ), a2_1(x, y, λ), and a2_0(x, y, λ), respectively, there are three undetermined coefficients and three passing points, so the exact solution is found as follows.

[0162] a2_2(x, y, λ) = ((Y1 - Y2) * (X1 - X3) - (Y1 - Y3) * (X1 - X2)) / ((X1 - X2) * (X1 - X3) * (X2 - X3))

[0163] a2_1(x, y, λ) = (Y1 - Y2) / (X1 - X2) - (a2_2) * (X1 + X2)

[0164] a2_0(x, y, λ) = Y1 - (a2_2) * X1 * X1 - (a2_1) * X1

[0165] As for the regression illumination correction coefficient, from the viewpoint of improving the robustness of the correction coefficient as well, a process of reducing the influence caused by measurement error on the correction coefficient calculated as described above can also be added. As one of the methods, a method of calculating a moving average value through a plurality of pixels adjacent to the pixel of which the coefficient is calculated can be cited. Below, as an example, is a correction coefficient in the case of calculating a moving average value through ±5 pixels adjacent to the pixel of which the coefficient is calculated.

[0166] [Equation 8]

[0167]

[0168] Using the regression illumination correction coefficient inherent to the pixel (x, y), that is, a2_2(x, y, λ) · a2_1(x, y, λ) · a2_0(x, y, λ), which is calculated as such, the measurement value of the measurement object 100, which is the measurement object, is corrected (regression illumination correction), and the influence of the regression illumination characteristics of the color measurement device 1 is removed.

[0169] When the reflectance before and after correction is set as Ref0(x, y, λ) and Ref rerefcorr (x, y, λ), respectively, the regression illumination characteristics can be approximated as a quadratic function, and can be corrected as follows, for example.

[0170] [Equation 9]

[0171] Ref rerefcorr (x, y, λ) = Ref0(x, y, λ) - (a2_2(x, y, λ) * Ref0(x, y, λ) ^ 2 + a2_1(x, y, λ) * Ref0(x, y, λ) + a2_0(x, y, λ))

[0172] Alternatively, as the regression illumination correction coefficient, a2_2′(x, y, λ)·a2_1′(x, y, λ)·a2_0′(x, y, λ) after moving average may be used.

[0173] Furthermore, the regression reflection correction coefficients a2_2(x, y, λ), a2_1(x, y, λ), and a2_0(x, y, λ) for each pixel may each be a continuous function, but can also be approximated by a polynomial such as a quadratic function. In this case, if the correction coefficient is calculated for only one pixel, the correction coefficients for the remaining pixels can be calculated based on the position coordinates of the measurement surface, and this correction coefficient can be used as the final correction coefficient. This has the advantage of not needing to store the correction coefficient for each pixel; only the correction coefficient for a single pixel needs to be stored.

[0174] In this way, it is believed that in the previous unevenness correction, although uniformity is ensured at the correction point, high correction accuracy cannot be ensured when the brightness is different, because of the spatial pixel dependence of the regression illumination characteristic. That is, if the previous unevenness correction is performed at the white point without implementing the regression illumination correction, or even if it is implemented, the same correction coefficient is used for all pixels, the absolute value accuracy cannot be fully maintained at the intermediate color and black due to the influence of the regression illumination characteristic of each pixel of the two-dimensional area sensor 133. As a result, it is speculated that it will appear as poor uniformity within the measurement surface. By using the regression illumination coefficient for correction, the influence of the regression illumination characteristic can be eliminated. In addition, it will not lead to unnecessary enlargement of the colorimetric device 1 or increase in cost, but can perform uniform measurement without error in each part within the measurement surface at any brightness (low brightness to high brightness) that the user wants to measure, and can achieve improved accuracy of the measurement result.

[0175] Figure 6 The graph shows the correction effect and is a graph showing the characteristics when unevenness correction and regression illumination correction are performed. Figure 6 The graph corresponds to Figure 8 The vertical and horizontal axes of the graphs (a) and (b) are Figure 8 In the diagram, S1 is a calibration white plate, and S2 to S5 are reference samples of achromatic colors with different brightness. Figure 8 (a) and (b) are the same.

[0176] from Figure 6 As can be clearly seen in the graphs shown, the baseline samples with varying brightness are significantly improved, with the error being eliminated across the entire measurement area by the regression illumination correction.

[0177] in addition, Figure 7is a graph showing the difference (standard deviation) of the measured values of each part of the measurement surface when the color measurement device 1 measures each part of the measurement surface in the SCI method for each of the samples S1 to S5 in the case where no correction is performed, only unevenness correction is performed, both unevenness correction and regression illumination correction are performed. The four bar graphs in each sample from the left represent σL * , σa * , σb * , and σdE.

[0178] As can also be understood from Figure 7 , it can be seen that by performing both unevenness correction and regression illumination correction, the measured values of each part of the measurement surface become uniform.

[0179] In addition, it is not necessary to perform the calculation of the unevenness correction coefficient and the regression illumination coefficient every time measurement is performed. The temporarily calculated unevenness correction coefficient and regression illumination coefficient can be stored, and the stored correction coefficients can be called out to perform correction at the time of measurement.

[0180] The above describes one embodiment of the present application, but the present application is not limited to the above-described embodiment. For example, although both unevenness correction and regression illumination correction are performed, only regression illumination correction can be performed.

[0181] In addition, although the unevenness correction coefficient and the regression illumination correction coefficient are calculated for each wavelength, they can be calculated for each of the tristimulus values XYZ.

[0182] In addition, the case where diffuse reflection light from the light source 12 is measured is described, but the case where transmitted light is measured can also be described.

[0183] This application claims priority from Japanese Patent Application No. 2023-029998 filed on February 28, 2023, the disclosure of which is incorporated herein by reference in its entirety.

[0184] The present application can be utilized as a color measurement device that measures the color, reflectance, and the like of an object.

[0185] BRIEF DESCRIPTION OF DRAWINGS 1 … color measurement device; 2 … personal computer (data processing device); 11 … integrating sphere (illumination section); 12 … light source (illumination section); 13 … light receiving section; 14 … arithmetic section; 15 … control section; 16 … recording section; 17 … display section; 18 … operation panel section; 19 … transceiver section; 21 … transceiver section; 100 … measurement object; 111 … inner surface; 112 … measurement opening; 113 … normal line; 114 … light receiving opening; 115 … light blocking plate; 131 … light splitting section; 132 … imaging lens; 133 … area sensor (photoelectric conversion section); 134 … pixel.

Claims

1. A colorimetric device, wherein: have: The lighting department is responsible for illuminating the measured object; a two-dimensional photoelectric conversion unit that splits light from a plurality of locations on the surface of the object being measured illuminated by the illumination unit into individual wavelengths, receives the light, and converts the received light into electrical signals; a calculation unit for calculating a regression illumination correction coefficient for correcting an error for each pixel caused by re-illumination in the colorimetric device based on light reception results of the photoelectric conversion unit for a plurality of samples having uniform luminance in respective portions of the measurement surface and different luminance across the measurement surface; as well as The correction unit corrects the measurement value of the measured object using the regression illumination correction coefficient calculated by the calculation unit.

2. The colorimetric device according to claim 1, wherein: The calculation unit calculates a nonuniformity correction coefficient for correcting nonuniformity of each pixel caused by components of the colorimetric device based on a result of light reception of the sample by the photoelectric conversion unit. The correction unit corrects the measurement value of the measured object using the unevenness correction coefficient calculated by the calculation unit.

3. The colorimetric device according to claim 1, wherein: The calculation unit calculates a regression illumination correction coefficient for each wavelength of the spectral analysis.

4. The colorimetric device according to claim 2, wherein: The calculation unit calculates the unevenness correction coefficient for each wavelength of the spectral distribution.

5. The colorimetric device according to claim 1, wherein: The calculation unit calculates the regression illumination correction coefficient according to each tristimulus value.

6. The colorimetric device according to claim 2, wherein: The calculation unit calculates the unevenness correction coefficient for each tristimulus value.

7. The colorimetric device according to claim 1, wherein: The above-mentioned sample is a sample in which the distribution of the measured values ​​at each wavelength is considered to be approximately uniform.

8. The colorimetric device according to claim 1, wherein: The calculation unit calculates the second-order coefficient a2_2(x, y, λ), the first-order coefficient a2_1(x, y, λ), and the zero-order coefficient a2_0(x, y, λ) as the regression illumination correction coefficient when the horizontal axis is approximated as a quadratic function with the horizontal axis set to the spectral reflectance and the vertical axis set to the deviation from the reference value using the following formula 1. The correction unit uses the calculated quadratic function of each regression illumination correction coefficient to correct the measured value according to the following formula 2: [Formula 1] a2_2(x,y,λ)=((Y1-Y2)*(X1-X3)-(Y1-Y3)*(X1-X2)) / ((X1-X2)*(X1-X3)*(X2-X3)) a2_1(x,y,λ)=(Y1-Y2) / (X1-X2)-(a2_2)*(X1+X2) a2_0(x,y,λ)=Y1-(a2_2)*X1*X1-(a2_1)*X1 Here, x and y represent the xy coordinates of the measurement surface, and λ represents the wavelength. For three samples, let the target measurement values ​​be Rc_1(λ), ​​Rc_2(λ), and Rc_3(λ), respectively; let the measurement values ​​of the colorimetric device to be calibrated be Rs_1(λ), ​​Rs_2(λ), and Rs_3(λ), respectively; let the regression illumination error be ΔRn=Rs_n(x,y,λ)-Rc_n(x,y,λ), and n=1,2,3, respectively. (X1,Y1)=(Rs_1(x,y,λ),ΔR1(x,y,λ)) (X2,Y2)=(Rs_2(x,y,λ),ΔR2(x,y,λ)) (X3,Y3)=(Rs_3(x,y,λ),ΔR3(x,y,λ)) [Formula 2] [Formula 10] Ref rerefcorr (x,y,λ)=Ref0(x,y,λ)-(a2_2(x,y,λ)×Ref0(x,y,λ)^2+a2_1(x,y,λ)×Ref0(x,y,λ)+a2_0(x,y,λ)) Among them, Ref0(x,y,λ) is the measurement value before correction, Ref rerefcorr (x,y,λ) is the corrected measurement value.

9. The colorimetric device according to claim 2, wherein: The above calculation unit sets the target measurement value to Count target (λ), and when the measured value of the second sample is set to Count(x, y, λ), the unevenness correction coefficient Mura(x, y, λ) is calculated using the following formula: [Math 11] The correction unit corrects the measured value Count(x, y, λ) to the corrected measured value Count by the following formula: corr (x,y,λ), [Mathematical formula 12] Conut corr (x,y,λ)=Free(x,y,λ)*Count(x,y,λ)。 10. The colorimetric device according to claim 1, wherein: The above samples contain gray or black blocks.

11. The colorimetric device according to claim 2, wherein: The above samples contain white plates.

12. The colorimetric device according to claim 1, wherein: After calculating the regression illumination correction coefficient for each pixel, the calculation unit calculates a weighted average value of adjacent pixels and sets the weighted average value as a final correction coefficient.

13. The colorimetric device according to claim 1, wherein: When the regression reflection correction coefficients of each pixel are continuous, the calculation unit approximates each regression illumination coefficient as a polynomial, calculates the coefficient of only one pixel, calculates the coefficients of all pixels according to the position coordinates of the measurement surface, and sets the coefficients of all pixels as the final correction coefficients.

14. The colorimetric device according to claim 2, wherein: After calculating the unevenness correction coefficients for each pixel, the calculation unit calculates a weighted average value of adjacent pixels and sets the weighted average value as a final correction coefficient.

15. The colorimetric device according to claim 2, wherein: After calculating the unevenness correction coefficient for only one pixel, the calculation unit calculates correction coefficients for all pixels based on the position coordinate relationship of known in-plane unevenness caused by components of the measurement device, and sets the correction coefficients for all pixels as final correction coefficients.

16. The colorimetric device according to any one of claims 1 to 15, wherein: The light from the plurality of positions on the surface of the object to be measured is reflected light from the plurality of positions.

17. The colorimetric device according to any one of claims 1 to 15, wherein: As the illumination unit, an integrating sphere is used that diffusely reflects light from a light source on its inner surface.

18. A data processing device, wherein: have: a receiving unit configured to receive, from a colorimetric device including an illumination unit and a two-dimensional photoelectric conversion unit, light reception results of a plurality of samples having uniform brightness at respective portions of a measurement surface and varying brightness across the measurement surface, the illumination unit illuminating the measurement object, the photoelectric conversion unit splitting light from a plurality of locations on the surface of the measurement object illuminated by the illumination unit into electrical signals for each wavelength; a calculation unit that calculates a regression illumination correction coefficient for correcting an error of each pixel caused by re-illumination in the colorimetric device based on the light reception result received by the receiving unit; as well as The correction unit corrects the measurement value of the measured object using the regression illumination correction coefficient calculated by the calculation unit.

19. A measurement correction method, wherein: include: an illumination step of illuminating, by an illumination unit, a plurality of samples having uniform brightness in respective portions of a measurement surface and different brightness on the measurement surface; a conversion step of splitting the light from a plurality of locations on the surface of the sample illuminated by the illumination unit into respective wavelengths, receiving the light through a two-dimensional photoelectric conversion unit, and converting the received light into an electrical signal; a calculation step of calculating a regression illumination correction coefficient for correcting an error of each pixel caused by re-illumination in the colorimetric device based on a result of light reception of the sample by the photoelectric conversion unit; as well as The correction step corrects the measured value of the measured object using the regression illumination correction coefficient calculated in the above calculation step.

20. A program for causing a computer to perform the following steps: a receiving step of receiving, from a colorimetric device comprising an illumination unit and a two-dimensional photoelectric conversion unit, light reception results of a plurality of samples having uniform brightness at respective portions of a measurement surface and varying brightness across the measurement surface, the illumination unit illuminating an object to be measured, the photoelectric conversion unit splitting light from a plurality of locations on a surface of the object to be measured illuminated by the illumination unit into electrical signals for each wavelength; a calculation step of calculating a regression illumination correction coefficient for correcting an error of each pixel caused by re-illumination in the colorimetric device based on the light reception result received in the receiving step; as well as The correction step corrects the measured value of the measured object using the regression illumination correction coefficient calculated in the above calculation step.

Images