WO2022220196A1

WO2022220196A1 - Correcting device, measuring instrument, correcting method, and program

Info

Publication number: WO2022220196A1
Application number: PCT/JP2022/017345
Authority: WO
Inventors: 健二井村; 克敏 ▲鶴▼谷
Original assignee: コニカミノルタ株式会社
Priority date: 2021-04-12
Filing date: 2022-04-08
Publication date: 2022-10-20
Also published as: KR20230154273A; JPWO2022220196A1; CN117157510A

Abstract

In a measuring instrument (100), light (6) to be measured is received by one or more sensors (x), (y), (z) having a spectral sensitivity approximating a target spectral sensitivity, and by a spectrograph (3), and a characteristic of the light to be measured is obtained on the basis of outputs from the sensors, wherein errors in the sensor outputs resulting from errors in the spectral sensitivity of the sensors (x), (y), (z) relative to the target spectral sensitivity are estimated from the target spectral sensitivity, the spectral sensitivity of each sensor, measured and stored in advance, and a spectral distribution of the light to be measured, measured by the spectrograph (3), and are corrected.

Description

Correction device, measuring instrument, correction method and program

The present invention provides a correction device, a measurement device, a correction method, and a correction device for correcting an output error caused by a spectral sensitivity error of a sensor used in a filter-type measurement device for measuring the luminance, chromaticity, etc. of radiated light of a display or the like. Regarding the program.

Measuring instruments that measure the luminance, chromaticity, etc. of radiated light from displays, etc. come in spectral and filter types, both of which obtain the tristimulus values of the light to be measured and convert them into luminance, chromaticity, etc.

A spectroscopic measuring instrument obtains tristimulus values by multiplying the spectral distribution of the measured light measured by a spectrograph composed of a diffraction grating and a high-sensitivity sensor array, and theoretical color matching functions. This spectroscopic type is in principle excellent in spectral sensitivity accuracy, but has disadvantages such as long measurement time, high cost, and large size.

On the other hand, a filter-type measuring instrument, which is also called a stimulus value direct reading type, receives the light to be measured with a sensor having a spectral sensitivity that approximates the color matching function, and directly obtains an output correlated with the tristimulus value. This filter-type measuring instrument surpasses spectroscopic-type measuring instruments in terms of measurement speed, and is superior in terms of size and cost. A disadvantage is that the application area is limited.

One of the major application fields of luminance meters is measurement for evaluation of display image quality. Spectral sensitivity accuracy comparable to that of high-precision spectroscopy is now required.

In addition, the quality evaluation of displays involves colorimetric measurements of many colored lights. For example, in gamma measurement (EOTF measurement), at least 25 levels of measurement are performed for the four colors B (blue), G (green), R (red), and W (white). As the performance of displays improves, the number of measurements per unit is increasing, and there is a demand for shortening the measurement time.

Therefore, it has been conventionally practiced to improve the accuracy of the sensor spectral sensitivity of filter-type measuring instruments, which have excellent measurement speed.

Many of the conventional techniques employ a matrix calibration method that reduces errors caused by spectral sensitivity errors using a calibration matrix obtained using a plurality of calibration reference lights. This method is effective when the light to be measured can be combined with the calibration reference light (when additive color mixture is established).

For example, in Patent Document 1, using a calibration matrix of tristimulus values obtained by using the primary lights (three lights of B, G, and R) of the display as the calibration reference lights, arbitrary colors synthesized by additive color mixture of the primary lights Techniques have been disclosed for improving the accuracy of the tristimulus values of the light to be measured.

When calibrating various types of displays with different spectral distributions of primary light using this technology, it is necessary to actually measure the tristimulus values of the primary light for each type and obtain a calibration matrix, which is time-consuming.

In

Patent Documents

2 and 3, the tristimulus values are obtained by numerical calculation by combining the color matching function and the pre-measured sensor spectral sensitivity of the filter-type luminance meter with the pre-obtained spectral distribution of the primary light of each type of display. By estimating , a technique is disclosed that eliminates the need to actually measure color values for each type.

In addition, in Patent Document 4, a specific position of the measurement area is measured by both the filter type and the spectrophotometer in the filter type two-dimensional colorimeter, and the correction coefficient obtained from the tristimulus values of both is applied to the entire measurement area. A technique has been proposed that aims to improve the accuracy of two-dimensional colorimetry by doing so.

JP-A-06-323910 JP 2012-215570 A Japanese Patent Application No. 2019-39795 JP-A-6-201472

However, the techniques of Patent Documents 1 to 3 are basically matrix calibration using the color values of the calibration reference light, and high accuracy is achieved only when the light to be measured can be synthesized with the calibration reference light (when additive color mixture is established). cannot be obtained, and there is a limit to suppressing the color value error that is caused by the sensor spectral sensitivity error and depends on the spectral distribution of the light to be measured.

Furthermore, in displays such as OLEDs (Organic Light Emitting Diodes), where the spectral distribution of radiated light is highly dependent on temperature, the difference in element temperature due to the difference in self-heating between low and high luminances causes the spectral distribution to changes, there is a problem that additive color mixture does not hold and matrix calibration does not work well.

In addition, in order to achieve the accuracy of the spectrophotometer level with the technique of Patent Document 4, the built-in spectrophotometer must have the accuracy of the spectrophotometer level. However, there is a problem that it is acceptable only in limited applications.

The present invention has been made in view of such a technical background. The purpose is to provide a correction device, a measuring device, a correction method, and a program that enable a filter type measuring device to measure any light to be measured with high accuracy and high speed by correcting based on the spectral distribution of and

The above objects are achieved by the following means.
(1) A measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and obtains the characteristics of the light to be measured based on the output of the sensors. , the error in the sensor output due to the spectral sensitivity error from the target spectral sensitivity of the sensor,
the target spectral sensitivity;
a pre-measured and stored spectral sensitivity of the sensor;
a spectral distribution of the light to be measured measured by the spectrograph;
A correction device that estimates and corrects from
(2) The correction device according to (1) above, wherein the spectrograph has a wavelength pitch and a half width of 4 nm or more.
(3) The correction device according to (1) or (2) above, wherein the permissible repeatability of the spectrograph is ten times or more the permissible repeatability of the measuring instrument.
(4) The spectral sensitivity error e _n (λ) of the spectral sensitivity s' _n (λ) of an arbitrary sensor n among the one or more sensors from the target spectral sensitivity s _n (λ) to e _n (λ)= Obtained by s' _n (λ)－s _n (λ),
As the spectral distribution I'(λ) of the light I to be measured measured by the spectrograph, the output error E _n of the sensor n caused by the spectral sensitivity error e _n (λ) is

estimated by
A simulated sensor output S' _n,sim is obtained from the spectral distribution I'(λ) of the light I to be measured measured by the spectrograph and the spectral sensitivity s' _n (λ) of the sensor n.

estimated by
The output error rate R _err,n when the light I to be measured is measured by the sensor n is calculated by R _err,n =E _n /S' _n,sim ,
4. The correction device according to any one of the preceding items 1 to 3, wherein the sensor output S' _n is corrected to the corrected sensor output S' _n,corr by S' _n,corr =S' _n ×(1−R _err,n ). .
(5) When the light to be measured consists of one or more monochromatic lights or ultra-narrow band lights,
Obtaining the wavelength and intensity of each of the monochromatic light or the ultra-narrow band light based on the narrow band spectrum of the monochromatic light or the ultra-narrow band light in the spectral distribution of the light to be measured measured by the spectrograph;
Obtaining the target spectral sensitivity and the sensor spectral sensitivity at the wavelength by interpolating the target spectral sensitivity and the sensor spectral sensitivity with respect to the wavelength;
5. The correcting device according to any one of the preceding items 1 to 4, wherein an error in the sensor output caused by the spectral sensitivity error is estimated based on the target spectral sensitivity at the wavelength, the sensor spectral sensitivity, and the intensity.
(6) where R is the wavelength range that includes the narrowband spectrum, and λ⊂R is the measurement wavelength within R, the wavelength _λc and the intensity A of monochromatic light or ultra-narrowband light are obtained by the following formula,

The spectral sensitivity error e _n (λ) is interpolated to obtain the spectral sensitivity error e _n (λ _c ) at the wavelength λ _c , and the sensor output error E _n is calculated by E _n =A· _en (λ _c ). presume,
6. The correcting device according to item 5, citing item 4, in which the simulated sensor output S' _n,sim is estimated by S' _n,sim =A·s' _n (λ _c ).
(7) The correcting device according to (5) or (6) above, further comprising detecting means for detecting whether the light to be measured is composed of one or more of monochromatic light and ultra-narrow band light.
(8) The correcting device according to any one of the preceding items 1 to 7, wherein the target spectral sensitivity is standard luminosity, and the measuring device is a luminance meter or an illuminance meter for measuring luminance or illuminance of the object to be measured.
(9) The correction device according to any one of the preceding items 1 to 7, wherein the target spectral sensitivity is a color matching function, and the measuring device is a color luminance meter or a color luminance meter for measuring the color characteristics of the object to be measured.
(10) A measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and obtains the characteristics of the light to be measured based on the output of the sensors. A measuring instrument comprising the correction device according to any one of the preceding items 1 to 9.
(11) A measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and obtains the characteristics of the light to be measured based on the output of the sensors. , the error in the sensor output caused by the spectral sensitivity error from the target spectral sensitivity of the sensor, the correction device,
the target spectral sensitivity;
a pre-measured and stored spectral sensitivity of the sensor;
a spectral distribution of the light to be measured measured by the spectrograph;
Correction method to estimate from and correct.
(12) The correction method according to (11) above, wherein the spectrograph has a wavelength pitch and a half width of 4 nm or more.
(13) The correction method according to (11) or (12) above, wherein the permissible repeatability of the spectrograph is 10 times or more the permissible repeatability of the measuring instrument.
(14) The spectral sensitivity error e _n (λ) of the spectral sensitivity s′ _n (λ) of any sensor n among the one or more sensors from the target spectral sensitivity s _n (λ) is e _n (λ)= Obtained by s' _n (λ)－s _n (λ),
Let I'(λ) be the spectral distribution of the light I to be measured measured by the spectrograph, and let the output error E _n of the sensor n caused by the spectral sensitivity error e _n (λ) be

estimated by
The output error rate R _err,n when the light I to be measured is measured by the sensor n is calculated by R _err,n =E _n /S' _n,sim ,
14. The correction method according to any one of the preceding items 11 to 13, wherein the sensor output S' _n is corrected to the corrected sensor output S' _n,corr by S' _n,corr =S' _n ×(1−R _err,n ). .
(15) When the light to be measured consists of one or more monochromatic lights or ultra-narrow band lights,
Obtaining the wavelength and intensity of each of the monochromatic light or the ultra-narrow band light based on the narrow band spectrum of the monochromatic light or the ultra-narrow band light in the spectral distribution of the light to be measured measured by the spectrograph;
Obtaining the target spectral sensitivity and the sensor spectral sensitivity at the wavelength by interpolating the target spectral sensitivity and the sensor spectral sensitivity with respect to the wavelength;
15. The correction method according to any one of the preceding items 11 to 14, wherein an error in the sensor output caused by the spectral sensitivity error is estimated based on the target spectral sensitivity at the wavelength, the sensor spectral sensitivity, and the intensity.
(16) Let R be the wavelength range that includes the narrowband spectrum, and λ⊂R be the measurement wavelength within R, and obtain the wavelength λ _c and the intensity A of monochromatic light or ultra-narrow band light by the following formula,

The spectral sensitivity error e _n (λ) is interpolated to obtain the spectral sensitivity error e _n (λ _c ) at the wavelength λ _c , and the sensor output error E _n is calculated by E _n =A· _en (λ _c ). presume,
16. The correction method according to item 15, which quotes item 14, in which the simulated sensor output S' _n,sim is estimated by S' _n,sim =A·s' _n (λ _c ).
(17) The correction method according to (15) or (16) above, which comprises detecting means for detecting whether the light to be measured is composed of one or more of monochromatic light and ultra-narrow band light.
(18) The correction method according to any one of the preceding items 11 to 17, wherein the target spectral sensitivity is standard luminosity, and the measuring device is a luminance meter or an illuminometer for measuring the luminance or illuminance of the object to be measured.
(19) The correction method according to any one of the preceding items 11 to 17, wherein the target spectral sensitivity is a color matching function, and the measuring device is a color luminance meter or a color luminance meter for measuring color characteristics of the object to be measured.
(20) A program for causing a computer to execute the correction method according to any one of (11) to (19) above.
(21) A measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and obtains characteristics of the light to be measured based on outputs of the sensors. There is
The sensor resulting from a spectral sensitivity error of the spectral sensitivity of the sensor from the target spectral sensitivity based on the spectral distribution of the light to be measured measured by the spectrograph, the target spectral sensitivity, and the sensor spectral sensitivity measured in advance. A measuring instrument that estimates and corrects for errors in the output of
(22) The measuring instrument according to (21) above, wherein the spectrograph has a wavelength pitch and a half width of 4 nm or more.
(23) The measuring device according to (21) or (22) above, wherein the permissible repeatability of the spectrograph is ten times or more the permissible repeatability of the measuring device.

According to the inventions described in the preceding paragraphs (1) and (11), the error in the sensor output due to the spectral sensitivity error from the target spectral sensitivity of the sensor of the filter type measuring device is corrected to the target without depending on matrix calibration. The spectral sensitivity, the sensor spectral sensitivity measured and stored in advance, and the spectral distribution of the light to be measured measured by the spectrograph are estimated and corrected, so that the light to be measured is synthesized by additive color mixture of the calibration reference light. Even if it is not possible, it is possible to perform highly accurate measurement using a filter-type measuring instrument. Moreover, high-speed measurement is possible without the need to use a highly accurate spectrograph.

According to the inventions described in the preceding paragraphs (2) and (12), the wavelength pitch and the half width of the spectrograph are 4 nm or more, so a high-precision spectrograph is unnecessary, and the cost of the spectrograph and the measurement equipment The cost can be reduced and high-speed measurement becomes possible.

According to the inventions described in the preceding paragraphs (3) and (13), the allowable repeatable error of the spectrograph is 10 times or more the allowable repeatable error of the measuring instrument, so a highly accurate spectrograph is unnecessary. and the cost of the measuring instrument can be reduced, and high-speed measurement becomes possible.

According to the inventions described in the preceding items (4) and (14), the sensor output error caused by the spectral sensitivity error from the target spectral sensitivity of the sensor of the filter-type measuring device is reliably estimated and corrected. be able to.

According to the inventions described in the preceding paragraphs (5) and (15), even when the light to be measured is composed of one or more monochromatic lights or ultra-narrow band lights, the error in the output due to the spectral sensitivity error of the sensor can be estimated and corrected.

According to the inventions described in the preceding paragraphs (6) and (16), even when the light to be measured consists of one or more monochromatic lights or ultra-narrow band lights, output errors due to spectral sensitivity errors of the sensor can be reliably estimated and corrected.

According to the inventions described in the preceding paragraphs (7) and (17), it is possible to detect that the light to be measured consists of one or more of monochromatic light and ultra-narrow band light.

According to the inventions described in the preceding paragraphs (8) and (18), the target spectral sensitivity is the standard luminosity factor, and in the luminance meter or illuminance meter for measuring the luminance or illuminance of the measurement target, the target spectral sensitivity of the sensor Errors in the sensor output due to spectral sensitivity errors from can be estimated and corrected.

According to the inventions described in the preceding paragraphs (9) and (19), the target spectral sensitivity is a color-matching function, and in the color luminance meter or color luminance meter for measuring the color characteristics of the object to be measured, the target spectral sensitivity of the sensor is Errors in sensor output due to spectral sensitivity error from sensitivity can be estimated and corrected.

According to the invention described in the preceding item (10), it is possible to estimate and correct the sensor output error caused by the spectral sensitivity error of the sensor from the target spectral sensitivity within the measuring instrument.

According to the invention described in the preceding item (20), the error in the sensor output due to the spectral sensitivity error from the target spectral sensitivity of the sensor of the filter-type measuring device is combined with the target spectral sensitivity and is measured and stored in advance. A computer can be caused to perform a process of estimating and correcting from the spectral sensitivity of the sensor and the spectral distribution of the light to be measured measured by the spectrograph.

According to the inventions described in the preceding items (21) to (23), the error in the sensor output due to the spectral sensitivity error from the target spectral sensitivity of one or more sensors having spectral sensitivities close to the target spectral sensitivity is reduced. Spectroscopic distributions used in computations to estimate and correct can be measured at high speed with less accurate, less costly spectrographs.

1 is a schematic configuration diagram of a filter-type measuring instrument according to one embodiment of the present invention; FIG. Spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ) and color matching functions s _x (λ), s _y (λ), s _z (λ) and , spectral sensitivity errors e _x (λ), e _y (λ), and e _z (λ). 2 is a flow chart showing a correction procedure performed by the filter-type measuring instrument of FIG. 1; Spectral sensitivity s' _x (λ), s' _y (λ), s' _z (λ) and spectral sensitivity error e of three sensors, sensor set a with small spectral sensitivity error and sensor sets b and c with large spectral sensitivity error 4 is a graph showing _x (λ), e _y (λ) and e _z (λ) together with color matching functions s _x (λ), s _y (λ) and s _z (λ); FIG. 5 is a graph showing errors in tristimulus values before and after correction of primary light of LCD (Liquid Crystal Display) simulated and measured using sensor set b in FIG. 4 ; FIG. FIG. 5 is a graph showing errors in tristimulus values before and after correction of the primary light of an LCD simulated using sensor set c of FIG. 4; FIG. 5 is a graph showing errors in tristimulus values before and after correction of primary light of an OLED (Organic Light Emitting Diode) simulated and measured using the sensor set b of FIG. 4; FIG. 5 is a graph showing errors in tristimulus values before and after correction of OLED primary light simulated using sensor set c of FIG. 4; FIG. 4 is a graph showing spectral distributions of primary light and white light of a typical LCD. 4 is a graph showing the spectral distribution of primary light and white light of a typical OLED; 1 is a graph showing the spectral distribution of a typical LD (Laser Display) emitted light. 2 is a graph showing the output of a spectrograph with a wavelength pitch and a half width of 4 nm for the spectral distribution of typical LD radiation. FIG. 10 is a schematic configuration diagram of a filter-type measuring instrument according to another embodiment of the present invention; FIG. 4 is a schematic configuration diagram of a filter-type measuring instrument according to still another embodiment of the present invention, where (A) is a side view and (B) is a view of FIG. A viewed from the right. FIG. 10, which shows still another embodiment of the present invention, is a schematic configuration diagram when the spectrograph is a spectrophotometer independent of the filter-type measuring instrument. FIG. 10, showing still another embodiment of the present invention, is a configuration diagram in which a filter-type measuring instrument, a spectrometer as a spectrograph, and a personal computer (PC) as a correction device are independent of each other. . FIG. 10, showing still another embodiment of the present invention, is a schematic configuration diagram in which the correction device is a PC independent of the filter-type measuring device.

Hereinafter, embodiments of the present invention will be described based on the drawings.

FIG. 1 is a schematic configuration diagram of a filter-type color luminance meter, which is an example of a filter-type measuring instrument 100 according to one embodiment of the present invention. In this embodiment, a correction device is built into the filter type colorimeter.

The filter-type measuring instrument 100 shown in FIG. , and the light to be measured 6 (light to be measured I) emitted from the light source 5 to be measured enters the incident end of the four-branch optical fiber bundle 2 via the lens system 1 . In the following description, the three filtered sensors n are also referred to as sensors x, y, z.

The 4-branch optical fiber bundle 2 is branched into 4 from the middle portion in the length direction, and distributes the light incident from the incident end to the 4 branch portions 21 . The sensors x, y, and z having spectral sensitivities s' _x (λ), s' _y (λ), and s' _z (λ) are provided at the output ends of the three branch portions 21 of the four-branch optical fiber bundle 2, respectively. The light emitted from the emitting end of each branching portion 21 is received by the sensors x, y, and z with filters. Light from the exit end of the remaining one branching portion 21 enters the spectrograph 3 .

The spectrograph 3 splits the incident light into wavelengths, and receives the split light at each pixel of the sensor array for each wavelength.

Arithmetic control unit 4 controls the entire measuring instrument and estimates and corrects the output error caused by the spectral sensitivity error from the target spectral sensitivity of each of the three sensors x, y, and z. Also functions as a device. The estimation and correction of the output error are performed as described below by the arithmetic control unit 4 operating according to the correction program.

That is, sensor outputs _S'x , S'y, and S'z are obtained from three types of sensors x, _y , and _z , and pixel signals pi ( _i : pixel number) of the sensor array obtained from the spectrograph 3 are converted into obtained spectral distribution I'(λ). Next, the pre-measured and stored sensor spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ), and the target spectral sensitivities previously given and stored as data, color matching _Corrected _tristimulus _values _S _' _{_} Convert to _x,corr , S'y, _corr , S'z _,corr . The sensor spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ) and color matching functions s _x (λ), s _y (λ), s _z (λ ) may be a storage unit (not shown) in the filter-type measuring instrument 100 or an external storage device. If it is stored in an external storage device, the sensor spectral sensitivity s' _x (λ), s' _y (λ), s' _z (λ) or color matching function s _x (λ ), s _y (λ), s _z (λ).
[1] Correction algorithm 1
This correction algorithm 1 is a basic algorithm for performing correction.
(1) Creation of correction coefficient by simulated measurement Spectral distribution I'(λ) of measured light 6 (measured light I) measured by spectrograph 3 and stored spectral sensitivities s of sensors x, y, and z From ' _x (λ), s' _y (λ), s' _z (λ), simulated sensor outputs S' _x,sim , S' _y,sim , S' _{z ,sim} is calculated.

The stored color matching functions s _x (λ), s _y (λ), s _z (λ) and the stored spectral sensitivities s' _x (λ), s' _y ( λ), s' _z (λ), the spectral sensitivity errors e _x (λ), e _y (λ), and e _z (λ) are calculated using (Equation 4) to (Equation 6). The spectral sensitivity errors e _x (λ), e _y (λ), and e _z (λ) may be calculated and stored in advance.

e _x (λ) = s' _x (λ) - s _x (λ) (Formula 4)
e _y (λ) = s' _y (λ) - s _y (λ) (equation 5)
e _z (λ) = s' _z (λ) - s _z (λ) (equation 6)
Color matching functions s _x (λ), s _y (λ), s _z (λ) and spectral sensitivities s' _x (λ), s' _y (λ), s' _z ( λ) and spectral sensitivity errors e _x (λ), e _y (λ), and e _z (λ) are shown in the graph of FIG.

Next, the spectral distribution I'(λ) of the light 6 to be measured measured by the spectrograph 3, and the spectral sensitivity errors e _x (λ), e _y (λ), Output errors E _x , E _y , and E _z of sensors x, y, and z are obtained from E _z (λ) using (Equation 7) to (Equation 9).

Next, the output errors E _x , E _y , and E _z of the sensors x, y, and z calculated by (Equation 7) to (Equation 9) and the simulated sensor output S calculated by (Equation 1) to (Equation 3) From ' _x,sim , S' _y,sim , S' _z,sim , the output error rate (sensor output error per unit output) R _err,x , R _err,y , R _err,z .

R _err,x =E _x /S' _x,sim (equation 10)
_Rerr,y = _Ey / _S'y,sim (Formula 11)
_Rerr,z = _Ez / _S'z,sim (equation 12)
(2) Correction of sensor output The sensor outputs S' _x , S' _y , and S' _z measured by the three types of sensors _x , y, and z are corrected using (Equation 13) to (Equation 15). _,corr , S' _y,corr , S' _z,corr . The corrected sensor output is converted into corrected tristimulus values by known arithmetic processing and output from the filter-type measuring instrument 100 .

S' _x,corr =S' _x ×(1−R _err,x ) (Formula 13)
S' _y,corr =S' _y ×(1−R _err,y ) (Formula 14)
S' _z,corr =S' _z ×(1−R _err,z ) (Formula 15)
As shown by (Equation 1) to (Equation 9), the accuracy of the output error rates R _err,x , R _err,y , and R _err,z is determined by the color matching functions s _x (λ), s _y (λ), It depends on the accuracy of s _z (λ), spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ) and spectral distribution I'(λ).

Since the color matching functions s _x (λ), s _y (λ), s _z (λ) are theoretical values, there is no error, and the spectral sensitivities s' _x (λ), s' _y (λ) of the sensor measured during manufacturing , s' _z (λ) can be obtained with the required accuracy by allocating necessary equipment (such as an irradiation monochromator) and time, so that it substantially depends on the accuracy of the spectral distribution I'(λ). However, since the output error rate is sufficiently smaller than 1 (R _err,x , R _err,y , R _err,z <<1), the corrected sensor output S' _x,corr , S ' _y,corr , S' has limited effect on _z,corr and therefore limited effect on the corrected tristimulus values.
(3) Accuracy required for spectrograph 3 (concrete example)
As described above, the spectral distribution I'(λ) of the light to be measured 6 measured by the spectrograph 3 is used only for estimating the error to be corrected. The effect on the corrected tristimulus values is limited even if there are measurement repetition errors and absolute value errors.

Specifically, when the error due to the spectral sensitivity error of the output of the filter type sensor 100 is 5%, even if the estimated correction amount has an error of 20%, the error of the corrected tristimulus value is 0.05. It stays at (5%) x 0.2 (20%) = 0.01 (1%).

(4) Repeatability required for spectrograph 3 E _rep,ttl is the allowable repeatability of the measuring instrument, and R _err,max is the maximum value of the output error rate (the ratio of the sensor output error due to the spectral sensitivity error to the sensor output). Then, if other error factors can be ignored, the repeat error E _rep,spe of the spectrograph 3 should satisfy E _rep,spe <E _rep,ttl /R _err,max .

Specifically, when E _rep,ttl =1% and R _err,max =5%, E _rep,spe is allowed up to 20%.

Since the maximum value of the output error rate R _err,max of the filter type is generally less than 10%, the repeatability error E _rep,spe of the spectrograph 3 is at least 10 times the allowable repeatability error E _rep,ttl of the measuring instrument ( 1/0.1) is acceptable.

Since a large repeatable error in the spectral distribution can be allowed, the time required for measuring the spectral distribution by the spectrograph 3 can be shortened. At least 10 times the permissible repeatability of stimulus value output means that, in principle, the measurement time is less than 1/10 ² of the time required to obtain tristimulus values from the spectral distribution like a spectrophotometer. do.

In addition, since a large repetitive error can be allowed in the spectral distribution I'(λ), the optical brightness (NA) and sensitivity of the spectrograph 3 can be lowered, resulting in cost and size reductions.

(5) Correction Procedure A correction procedure executed by the filter-type measuring instrument 100 of FIG. 1 is shown in the flow chart of FIG.

Measurement of the light 6 to be measured by the three sensors x, y, z and the spectrograph 3 is started at the same time, and the sensor output S'n ( _n =x, _y , z) and the pixel output pi are obtained (step #1 and step #2).

After both measurements are completed, the arithmetic control unit 4 converts the pixel output p _i into the spectral distribution I'(λ) (step #3). Further, the sensor output error E _n is estimated from the spectral distribution I′(λ) and the spectral sensitivity error e _n (λ) using (Equation 1) to (Equation 12), and the output error rate R _err,n is obtained (step #4).

Next, the output error rate R _err,n is applied to the sensor output S' _n according to (Equation 13) to (Equation 15) to obtain the corrected sensor output S' _n,corr , which is converted to the corrected tristimulus value and output. (Step #5).

When the permissible repeat error E _rep,ttl of the measuring instrument is obtained for each of the sensors x, y, z and the spectrograph 3, the measurement time T _spe of the spectrograph 3 in step #2 is obtained from the sensors x, y, ttl in step #1. Although the measurement time for z is much longer than T _fil (typically 2 s versus 0.05 s), in the present embodiment, which allows for large repeatability errors in the spectrograph 3, the measurement time is reduced to T' _spe ≈ T _fil . , the total measurement time E _rep,ttl can be set to the level of the filter-type colorimeter (Since the calculation time of steps #3 to #5 can be ignored, the total measurement time T _ttl ≈ T' _spe ≈ T _fil ). .

Specifically, if the output error rate R _err,n (n=x, y, z) is 5% at maximum, the repeatable error E _rep,spe is magnified to 20 times (1/0.05) the allowable repeatability E _rep,ttl of the measuring instrument.

On the other hand, if the measurement time T' _spe ≈ 2 s of the spectrograph 3 is shortened to 0.05 s, which is the same as T _fil , the repetition error of the spectral distribution I'(λ) becomes √(2/0.05) = 6.3 times, but it is acceptable. It is sufficiently smaller than the enlargement ratio (20 times) of the repeat error E _rep,spe , and is not a problem.

In this case, spectrograph 3 still has a margin of more than 3 times (20 times/6.3 times) in the allowable repeatability error E _rep,spe , so trade it with spectrograph 3's optical brightness (NA) and sensitivity. As a result, the size and cost of the spectrograph 3 can be reduced.

(6) Wavelength pitch and half width of spectrograph 3 The accuracy of correction according to this embodiment depends on the accuracy of spectral distribution I'(λ), and therefore the wavelength pitch of spectrograph 3 that measures I'(λ) and Although it depends on the half-value width, below we will confirm their influence by simulated measurement of LCD and OLED.

4(a), 4(b), and 4(c) show the spectral sensitivities s' of the sensors x, y, and z of the sensor set a with a small spectral sensitivity error and the sensor sets b and c with a large spectral sensitivity error. _x (λ), s' _y (λ), s' _z (λ) and spectral sensitivity errors e _x (λ), e _y (λ), e _z (λ) are converted into color matching functions s _x (λ), It is a graph shown with s _y (λ) and s _z (λ). In FIGS. 4(a), 4(b), and 4(c), the left vertical axis is the spectral sensitivity s' _x (λ), s' _y (λ), s' _z (λ) and the color matching function s Values of _x (λ), s _y (λ), and s _z (λ) are shown, and the right vertical axis represents values of spectral sensitivity errors e _x (λ), e _y (λ), and e _z (λ).

Tristimulus values of LCD primary lights B, G, and R (spectral distribution is shown in FIG. 9) simulated and measured according to formulas (1) to (12) using sensor set b and sensor set c. 5 and 6 show the errors (absolute values) |ΔX|, |ΔY|, and |ΔZ| before and after the correction of . Similarly, FIGS. 7 and 8 show errors (absolute values) before and after correction of the tristimulus values of the OLED primary light (the spectral distribution is shown in FIG. 10) simulated and measured. The three bars for each error are the B, G, and R values from left to right.

The spectral distribution used for correction in the simulated measurement was measured with an isosceles triangular slit function and six types of spectrographs 3 having wavelength pitches and half-value widths shown in Table 1.

As shown in FIGS. 5 to 8, when the wavelength pitch and half width of Spectrograph 3 are 4 to 12 nm, the corrected errors |ΔX|, |ΔY|, and |ΔZ| are improved from the errors |ΔX _uc |, |ΔY _uc |, |ΔZ _uc |. In particular, the correction effect is remarkable at 4 to 8 nm, and the spectral sensitivity accuracy is set to the level of a spectral radiance meter, and the amount of light incident on each pixel of the spectrograph 3 is approximately 4 ² to 8 ² times that when both the wavelength pitch and the half value width are 1 nm. can be

In other words, since a large spectral distribution error can be tolerated in this embodiment, the wavelength pitch and half width of the spectrograph 3 are increased to increase the amount of light incident on each pixel, thereby further shortening the measurement time. It is also possible to trade the increase in light intensity with the optical brightness (NA) and sensitivity of the spectrograph 3 to reduce cost and size.
[2] Correction algorithm 2
This correction algorithm is for the emitted light of a laser display (LD).

The finite half-value width of the spectrograph 3 does not cause a large error in the radiant light from LCDs and OLEDs, which have broad spectral distributions as shown in FIGS. ) into the narrowband spectrum shown in FIG. 12, resulting in non-negligible errors. In other words, the actual sensor outputs S' _x , S' _y , S' _z from the laser light include the spectral sensitivities s' _x (λ _L ), s' _y (λ _L ₎ , s' _z (λ _L ), but simulated sensor outputs S′ _x,sim , S′ _y,sim , which are simulated and measured by (Equation 1) to (Equation 3) using the spectral distribution I′(λ) of FIG. The spectral sensitivity at wavelengths around the laser wavelength λ _L also contributes to S' _z,sim , resulting in an error.

Similarly, only the spectral sensitivity errors e _x (λ _L ), e _y (λ _L ), and e _z (λ _L ) at λ _L contribute to the actual sensor output error, but I′(λ) in FIG. The spectral sensitivity errors at wavelengths around the laser wavelength λ _L contribute to the sensor output errors E _x , E _y , and E _z estimated by (Equation 7) to (Equation 9) using The larger the half-value width, the larger the error).

In order to avoid this, in the correction algorithm 2, the steps (formula 1) to (formula 9) of the correction algorithm 1 are replaced with the following steps.
(1) Basic process (when the light to be measured consists of one LD radiation light)
First, in the spectral distribution I′(λ) of the LD radiation light obtained by the spectrograph 3, let R be the wavelength region including the narrowband spectrum (see FIG. 12), and the centroid of the wavelength region R given by the formula (16) The laser wavelength is approximated by the wavelength _λc , and the laser intensity is approximated by the integral intensity A given by the equation (17). In equations (16) and (17), λ⊂R represents all measured wavelengths in R. The wavelength range R can be, for example, a range four times the half width w of the spectrograph 3 (±2w) around the peak wavelength in the spectral distribution I′(λ).

_Sensor _spectral _sensitivity s' _x ₍ λ _c ), s' _y (λ _c ), s' _z (λ _c ), and estimate the simulated sensor output S' _x,sim , S' _y,sim , S' _z,sim by (Equation 18) .

Spectral sensitivity errors e _x (λ _c ), e _y (λ _c ) at centroid wavelength λ _c by interpolating stored spectral sensitivity errors e _x (λ), e _y (λ), e _z (λ) , e _z (λ _c ) are obtained, and the sensor output errors E _x , E _y , E _z are estimated by (Equation 19).

(Equation 18) and (Equation 19) have no wavelength components other than the centroid wavelength λ _c , and the simulated sensor outputs S′ _x,sim , S′ _y,sim , S′ _z,sim also have sensor output errors E _x , E _y and E _z also have no contribution of wavelength components other than λ _c . In this embodiment as well, the influence of the error in the spectral distribution of the laser light (specifically, the centroid wavelength λ _c and the intensity A) on the corrected sensor output is very small.
(2) Practical process (when the light to be measured consists of B, G, and R LD radiation)
As shown in FIG. 11, when the light to be measured includes the three primary lights _Ib , _Ig , and _Ir of the LD, the measured spectral distribution I'(λ) of each primary light is narrow as shown in FIG. It has a band spectrum. The centroid wavelengths λ _b , λ _g , λ _r and the integrated intensities A _b , _Ag , _Ar of each narrowband spectrum are obtained in the same manner as described above, and the spectral sensitivities s' _x (λ), s' _y (λ ), s' _z (λ) to obtain sensor spectral sensitivities s' _x (λ _b ), s' _y (λ _g ), s' _z (λ _r ) at centroid wavelengths λ _p , λ _g , and λ _r , … , s' _z (λ _b ), s' _z (λ _g ), s' _z (λ _r ) are obtained, and simulated sensor outputs S' _x,sim , S' _y,sim , S ' Estimate _z,sim .

_Spectral sensitivity errors e _x ₍ _λ ₎ , _e _x ₍ _λ _g ), e _x (λ _r ), … , e _z (λ _b ), e _z (λ _g ), e _z (λ _r ) are obtained, and the sensor output errors E _x , E _y , E Estimate _z .

(3) Discrimination of LD synchrotron radiation (ultra-narrow band light) The above process can be applied not only to laser light but also to ultra-narrow band light. ) is used. Whether or not the light to be measured is laser light or ultra-narrow band light can be determined by the operator, or can be automatically determined by the arithmetic control unit from the spectral distribution I'(λ). For example, the index T(Δ) shown in (Equation 22) is the correlation between I'(λ) and I'(λ+Δ) obtained by shifting I'(λ) by Δnm wavelength, and the self of I'(λ). For example, when Δ=16 nm, T(16)=0.00 in the spectral distribution of FIG. 12, but T(16)=0.00 in the spectral distributions of FIGS. 9 (LCD) and 10 (OLED). 0.80 and 0.72. If T is smaller than a threshold value (for example, 0.1), the arithmetic control unit 4 determines that the light is laser light or ultra-narrow band light.

FIG. 13 is a schematic configuration diagram of a filter-type measuring instrument 100 according to another embodiment of the invention. This embodiment differs from the filter-type measuring instrument shown in FIG.

The filter-type measuring instrument 100 of FIG. 13 includes a light pipe 22 into which light 6 to be measured from a light source 5 to be measured is incident via a lens system 1 consisting of a plurality of lenses, and a light pipe 22 connected to the output end face of the light pipe 22 . It is equipped with the resin fiber 23 of this. Sensors x, y, and z are arranged at the emission ends of three resin fibers 23 among the four resin fibers 23, and the spectrograph 3 is arranged at the emission end of one resin fiber 23. there is Then, the luminous flux incident on the light pipe 22 is distributed to four resin fibers 23 at the output end face, and the luminous flux distributed to the three resin fibers 23 is received by each sensor x, y, z, and one resin The spectrograph 3 receives the beam distributed to the fiber 23 .

14A and 14B are schematic configuration diagrams of a filter-type measuring instrument 100 according to still another embodiment of the present invention, where (A) is a side view and (B) is a view of (A) viewed from the right.

In this embodiment, the light 6 to be measured from the light source 5 to be measured is diffused by a dome-shaped diffusion plate 7, and the diffused light is received by three sensors x, y, and z arranged around a central lens 8. At the same time, the diffused light converged by the lens 8 is received by the spectrograph 3 .

FIG. 15 shows yet another embodiment of the invention. The filter-type measuring instrument 100 of FIGS. 1, 13 and 14 incorporates three sensors x, y, z, a spectrograph 3, and an arithmetic control section 4 functioning as a correction device. but
A filter type measuring instrument 100 of FIG. 15 incorporates three sensors x, y, z and an arithmetic control unit 4, and is connected to an independent spectrophotometer 31 as a spectrograph 3 and the like. In this embodiment, the light 6 to be measured is measured by three sensors x, y, z and an independent spectrophotometer 31, and these outputs are input to the arithmetic control unit 4, and the three sensors x, y , Estimate and correct the error in the output of z.

Further, as shown in FIG. 16, a filter type color luminance meter etc. as sensors x, y, z, a spectral luminance meter 31 etc. as a spectrograph 3, and a personal computer (PC) 41 etc. as a correction device are respectively may be independent. In this embodiment, the PC 41 receives the output of each sensor x, y, z of the filter-type luminance meter 100 and the output of the spectral luminance meter 31 via a network or the like, and Estimate and correct the errors in the outputs of the sensors x, y, z.

Further, as shown in FIG. 17, the three sensors x, y, z and the spectrograph 3 are incorporated in the filter-type measuring instrument 100, and the correction device is configured by an external PC 41 or the like. An external PC 41 may receive the output of each sensor x, y, z and the output of the spectrograph 3 via a network or the like to correct the sensor output.
As described above, the technique of this embodiment can
Using an inexpensive spectrograph with a wavelength pitch and a half-value width of 4 nm or more, preferably 4 to 8 nm, and a permissible repeatability error of 10 times or more than the permissible repeatability error of a measuring instrument, correction can be performed at high speed and with high accuracy.

Although the filter-type measuring instrument 100 having three sensors x, y, and z has been described in the above embodiment, the number of sensors is not limited to three. For example, colorimeters generally have three types of sensors for the color matching functions _x , y, and z. or a luminance meter with one type of sensor that approximates the standard luminous efficiency V(λ).

As described above, in this embodiment, the target spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ) of the sensors x, y, and z of the filter type measuring instrument 100 are The error of the sensor output due to the spectral sensitivity error from s _x (λ), s _y (λ), s _z (λ) is defined as the target spectral sensitivity s _x (λ), s _y (λ), s _z (λ ), sensor spectral sensitivities s' _x (λ), s' _y (λ), s' _z (λ) measured and stored in advance, and spectral distribution I' ( λ) and corrected. Unlike matrix calibration, highly accurate measurement is possible even with light to be measured that cannot be synthesized with calibration reference light. Moreover, high-speed measurement is possible without the need to use a highly accurate spectrograph.

This application claims the priority of Japanese Patent Application No. 2021-067207 filed on April 12, 2021, and the disclosure thereof constitutes a part of this application as it is. .

The present invention can be used as a correction device that corrects output errors caused by spectral sensitivity errors of sensors used in filter-type measuring instruments that measure the luminance, chromaticity, etc. of radiated light such as displays.

REFERENCE SIGNS LIST 1 lens system 2 4-branch optical fiber bundle 3 spectrograph 4 arithmetic control unit 5 measurement object 6 light I to be measured
7 diffusion plate 8 center lens 21 branching part 22 light pipe 23 resin fiber 31 spectral luminance meter 41 correction device (personal computer)
100 filter type measuring instrument

Claims

The sensor of a measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities approximating a target spectral sensitivity and a spectrograph, and determines characteristics of the light to be measured based on outputs of the sensors. The sensor output error due to the spectral sensitivity error from the target spectral sensitivity of the spectral sensitivity of
the target spectral sensitivity;
a pre-measured and stored spectral sensitivity of the sensor;
a spectral distribution of the light to be measured measured by the spectrograph;
A correction device that estimates and corrects from
The correction device according to claim 1, wherein the spectrograph has a wavelength pitch and a half width of 4 nm or more.
The correction device according to claim 1 or 2, wherein the permissible repeatability of the spectrograph is ten times or more the permissible repeatability of the measuring instrument.
The spectral sensitivity error e n (λ) of the spectral sensitivity s' n (λ) of any sensor n among the one or more sensors from the target spectral sensitivity s n (λ) is e n (λ)=s' n (λ)－s n (λ),
As the spectral distribution I'(λ) of the light I to be measured measured by the spectrograph, the output error E n of the sensor n caused by the spectral sensitivity error e n (λ) is

estimated by
A simulated sensor output S' n,sim is obtained from the spectral distribution I'(λ) of the light I to be measured and the spectral sensitivity s' n (λ) of the sensor n.

estimated by
The output error rate R err.n when the light I to be measured is measured by the sensor n is calculated by R err.n =E n /S' n,sim ,
The correction according to any one of claims 1 to 3, wherein the sensor output S' n is corrected to the corrected sensor output S' n, corr by S' n, corr = S' n ×(1-R err.n ). Device.
When the light to be measured consists of one or more monochromatic lights or ultra-narrow band lights,
Obtaining the wavelength and intensity of each of the monochromatic light or the ultra-narrow band light based on the narrow band spectrum of the monochromatic light or the ultra-narrow band light in the spectral distribution of the light to be measured measured by the spectrograph;
Obtaining the target spectral sensitivity and the sensor spectral sensitivity at the wavelength by interpolating the target spectral sensitivity and the sensor spectral sensitivity with respect to the wavelength;
The correction device according to any one of claims 1 to 4, wherein an error in the sensor output caused by the spectral sensitivity error is estimated based on the target spectral sensitivity at the wavelength, the sensor spectral sensitivity, and the intensity.
Let R be the wavelength range that includes the narrowband spectrum, and λ ⊂ R be the measurement wavelength within R, and obtain the wavelength λ c and the intensity A of monochromatic light or ultra-narrow band light by the following formula,

The spectral sensitivity error e n (λ) is interpolated to obtain the spectral sensitivity error e n (λ c ) at the wavelength λ c , and the sensor output error E n is calculated by E n =A· en (λ c ). presume,
6. A correction device according to claim 4, wherein the simulated sensor output S'n,sim is estimated by S'n,sim = A.s'n ( λc ).
7. The correction device according to claim 5 or 6, comprising detection means for detecting whether the light to be measured consists of one or more of monochromatic light and ultra-narrow band light.
The correction device according to any one of claims 1 to 7, wherein the target spectral sensitivity is standard luminosity, and the measuring device is a luminance meter or an illuminance meter for measuring luminance or illuminance of a measurement target.
The correction device according to any one of claims 1 to 7, wherein the target spectral sensitivity is a color matching function, and the measuring device is a color luminance meter or a color luminance meter for measuring the color characteristics of the object to be measured.
Light to be measured is received by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and the characteristics of the light to be measured are determined based on the output of the sensors, A measuring instrument comprising a correction device according to any one of claims 1-9.
The sensor of a measuring instrument that receives light to be measured by one or more sensors having spectral sensitivities approximating a target spectral sensitivity and a spectrograph, and determines characteristics of the light to be measured based on outputs of the sensors. The correction device corrects the sensor output error caused by the spectral sensitivity error from the target spectral sensitivity of
the target spectral sensitivity;
a pre-measured and stored spectral sensitivity of the sensor;
a spectral distribution of the light to be measured measured by the spectrograph;
Correction method to estimate from and correct.
The correction method according to claim 11, wherein the wavelength pitch and half width of the spectrograph are 4 nm or more.
The correction method according to claim 11 or 12, wherein the permissible repeatability of the spectrograph is 10 times or more the permissible repeatability of the measuring instrument.
The spectral sensitivity error e n (λ) of the spectral sensitivity s' n (λ) of any sensor n among the one or more sensors from the target spectral sensitivity s n (λ) is e n (λ)=s' n (λ)－s n (λ),
Let I'(λ) be the spectral distribution of the light I to be measured measured by the spectrograph, and let the output error E n of the sensor n caused by the spectral sensitivity error e n (λ) be

estimated by
A simulated sensor output S' n,sim is obtained from the spectral distribution I'(λ) of the light I to be measured measured by the spectrograph and the spectral sensitivity s' n (λ) of the sensor n.

estimated by
The output error rate R err,n when the light I to be measured is measured by the sensor n is calculated by R err,n =E n /S' n,sim ,
The correction according to any one of claims 11 to 13, wherein the sensor output S' n is corrected to the corrected sensor output S' n,corr by S' n,corr =S' n ×(1−R err,n ). Method.
When the light to be measured consists of one or more monochromatic lights or ultra-narrow band lights,
Obtaining the wavelength and intensity of each of the monochromatic light or the ultra-narrow band light based on the narrow band spectrum of the monochromatic light or the ultra-narrow band light in the spectral distribution of the light to be measured measured by the spectrograph;
Obtaining the target spectral sensitivity and the sensor spectral sensitivity at the wavelength by interpolating the target spectral sensitivity and the sensor spectral sensitivity with respect to the wavelength;
The correction method according to any one of claims 11 to 14, wherein an error in the sensor output caused by the spectral sensitivity error is estimated based on the target spectral sensitivity at the wavelength, the sensor spectral sensitivity, and the intensity.
Let R be the wavelength range that includes the narrowband spectrum, and λ ⊂ R be the measurement wavelength within R, and obtain the wavelength λ c and the intensity A of monochromatic light or ultra-narrow band light by the following formula,

The spectral sensitivity error e n (λ) is interpolated to obtain the spectral sensitivity error e n (λ c ) at the wavelength λ c , and the sensor output error E n is calculated by E n =A· en (λ c ). presume,
16. A correction method according to claim 15, wherein the simulated sensor output is estimated by S'n,sim =A· s'n ( λc ).
The correction method according to claim 15 or 16, further comprising detecting means for detecting that the light to be measured consists of one or more of monochromatic light and ultra-narrow band light.
The correction method according to any one of claims 11 to 17, wherein the target spectral sensitivity is standard luminosity, and the measuring device is a luminance meter or an illuminometer for measuring the luminance or illuminance of the object to be measured.
The correction method according to any one of claims 11 to 17, wherein the target spectral sensitivity is a color matching function, and the measuring device is a color luminance meter or a color luminance meter for measuring the color characteristics of the object to be measured.
A program for causing a computer to execute the correction method according to any one of claims 11 to 19.
Light to be measured is received by one or more sensors having spectral sensitivities close to a target spectral sensitivity and a spectrograph, and the characteristics of the light to be measured are determined based on the output of the sensor,
The sensor resulting from a spectral sensitivity error of the spectral sensitivity of the sensor from the target spectral sensitivity based on the spectral distribution of the light to be measured measured by the spectrograph, the target spectral sensitivity, and the sensor spectral sensitivity measured in advance. A measuring instrument that estimates and corrects for errors in the output of
The measuring instrument according to claim 21, wherein the spectrograph has a wavelength pitch and a half width of 4 nm or more.
23. The measuring device according to claim 21 or 22, wherein the permissible repeatable error of the spectrograph is 10 times or more the permissible repeatable error of the measuring device.