KR20070010812A - Color vision modeling and color correction method and system using color vision inspection - Google Patents

Abstract

The present invention relates to a method for describing the characteristics of standardized color vision deficiency (color vision deficiency) in the equipment for displaying color by using the results of the existing color vision test, the presence, type and degree of color vision abnormality The error score of the Faransworth-Munsell 100 Hue test (hereinafter referred to as FM100H test) and computerized FM100H test (hereinafter referred to as CFM100H test), which is one of the representative color vision test used to test It aims at the method of obtaining a correction value for correction, and correcting a color using this value.

In order to obtain a color correction value in such a color display equipment,

(1) modeling the spectral cone sensitivity functions of the color weakener according to the degree of deficiency of color vision,

(2) computer simulation of the color recognition characteristics of the color weakness using the spectral cone cell response function of the modeled color weakness,

(3) In modeling the color correction value by the existing color vision test method,

(3-1) matching the confusion line of the color weakness through the color weakness simulation of the color of the cap of the CFM100H test;

(3-2) Through the color weakness simulation of the color of the cap of the CFM100H test, the confusion region of the color weakening is widened as the degree of color vision abnormality increases,

(3-3) Through the color weakness simulation of the color of the cap of the CFM100H test, by measuring the perceptual color difference between the color weakness simulated color and the original color, the color discrimination ability according to the increase of the degree of color vision abnormality It characterized in that it comprises the step of measuring the degree of reduction of,

(4) The total error score of these two tests (total error score, TES) is performed by performing two types of color weakened CFM100H test and color-corrected CFM100H test (hereinafter, c-CFM100H test). Statistically modeling the degree of severity of color vision abnormality according to the total number of false scores of the CFM100H test,

(5) using the results, characterized in that the step of modeling the relationship between the total score of the CFM100H test and the degree of color vision abnormality severity,

(6) expressing a numerical value of color vision correction to provide a standard in color display equipment by using the relationship between the total score of the CFM100H test and the degree of color vision abnormality severity.

Description

Color vision modeling and color correction method and system using color vision inspection {METHOD AND APPARATUS OF COLOR VISION DEFICIENCY MODELING USING COLOR VISION TESTS AND COLOR COMPENSATION}

1 is a block diagram illustrating a procedure of a method of obtaining a standard value for generating a color vision abnormality for a color display device using a color vision test result according to an exemplary embodiment of the present invention.

Figure 2 is the degree of movement of the cone tip peak response proposed in the present invention

Spectral cone cell response function measured in the experiment

Fig. 4 The degree of movement of the cone response at each wavelength measured from the spectral cone response function of the color weak

Fig. 5 Spectral cone cell response function of modeled color weakness

6 is a flowchart illustrating a process of obtaining a color drug degree and color discrimination relationship from the spectral cone cell response function of a modeled color weakness

Fig. 7 Simulation process of color recognition characteristics of color weakness using spectral cone cell response function of color weakness modeled

Figure 8 shows an simulated color in the x-y color space for the color of the cap of the CFM100H test. Figure 8-a cap color distribution seen by (a) normal people, (b) red color abbreviation seen by 0.1 (b). (c) Cap color distribution with a red abbreviation of 0.9. FIG. 8-B (a) The cap color distribution seen by a normal person, (b) The cap color distribution seen by a green weak person of 0.1. (c) Cap color distribution with green abbreviation of 0.9.

Fig. 9 shows an example of chromaticity simulating the color visible by the color weaker for the color of the cap of the CFM100H inspection. Fig. 9-A: Chromaticity seen by the red weak. Fig. 9-B: Chromaticity seen by the green weak

10 The cognitive color difference between the color of the cap of the CFM100H test and its color simulated color is measured in the CIE L * a * b * color space. Fig. 10-A: Red abbreviation. Fig. 10-B: When the green abbreviation is. Figure 10-C: both red and green.

11 is a flowchart illustrating a color correction process in a color display device for improving color recognition ability of a color weak person;

12 is a flowchart illustrating a procedure for comparing color error scores and color weakness dimensions using the CFM100H test and the c-CFM100H test.

Fig. 13 t-test results of the mean of the total scores of the c-CFM100H test performed on the color weak

14 is a diagram showing the relationship between the estimated color score and the error score for the total score of the CFM100H test measured in the color weak

Figure 15 Linear regression model results of the amount of color weakness compared to the total score of the CFM100H test

Fig. 16 is a comparison diagram showing the relationship between the color difference estimation function and the color difference in the color recognition simulation of the color weakness through the cone cell response function modeling of the color weakness.

Recently, in order to improve the color recognition ability of color-impaired people, the study of color-recognized people through color display devices has become very important. Among them, the World Wide Web Consortium (W3C), the world's standard organization for Internet-based World Wide Web services, provides some guidelines for creating web content for the visually impaired. However, while these guidelines provide detailed technical details, they do not provide specific alternatives, such as how to create efficient designs and reusable processes.

Microsoft is also proposing some things to check in order to take into account color vision deficiencies when writing software. However, this is only a recommendation to be followed during the content or software production stage. Numerous content and software have not been produced in accordance with such recommendations, and content is still being produced regardless of this. In addition, due to technical limitations in content production such as real-time video, color correction services for color vision impaired are not available by prior research methods.

Conventional techniques for helping color-impaired people to better distinguish colors in everyday life include the use of color glasses that filter colors in specific wavelength ranges. Color-impaired people wear color glasses to see objects, so they can better distinguish colors. However, this method has a problem that the color vision impaired person looks different from the normal color, and it is difficult to consider the various severity of color pills.

By properly correcting the color perceived by a user with color vision impairment in a user terminal, a method of accurately recognizing color information of content has become important for the spread of color display apparatuses. Recently, although a fundamental treatment for color vision abnormalities is impossible, studies have been conducted to improve color recognition ability of color weakening by correcting colors emitted through display devices [Refs. 1 and 2].

In addition, international standardization of color blindness characteristics through MPEG-21 digital item adaptation descriptors has been conducted. However, the method of obtaining the standardized color vision abnormality has not been studied yet. In general, color pills have a wide variety of severity levels, and in order to correct colors more accurately, it is necessary to accurately determine the severity of color pills.

Characteristics of color vision abnormalities such as type and degree of color vision abnormalities are generally diagnosed by Pseudoisochromatic tests such as Ishihara test or HRR test. However, Farnsworth Munsell test or anomalouscope test is used to diagnose more precise color vision abnormality.

An object of the present invention is to show a method of accurately diagnosing color vision abnormality and to quantify the results of general-purpose Farnsworth Munsell tests to obtain a standardized color vision abnormality, and a color correction method based thereon.

Compared with the prior art, the features of the present invention are as follows.

(1) modeling the spectral cone sensitivity functions of the color weakener according to the degree of deficiency of color vision,

(2) computer simulation of the color recognition characteristics of the color weakness using the spectral cone cell response function of the modeled color weakness,

(3) In modeling the color correction value by the existing color vision test method,

(3-1) matching the confusion line of the color weakness through the color weakness simulation of the color of the cap of the CFM100H test;

(3-2) Through the color weakness simulation of the color of the cap of the CFM100H test, the confusion region of the color weakening is widened as the degree of color vision abnormality increases,

(3-3) Through the color weakness simulation of the color of the cap of the CFM100H test, by measuring the perceptual color difference between the color weakness simulated color and the original color, the color discrimination ability according to the increase of the degree of color vision abnormality It characterized in that it comprises the step of measuring the degree of reduction of,

(4) The total error score (TES) of these two tests is performed by performing two types of color-compensated CFM100H test and color-corrected CFM100H test (hereinafter, c-CFM100H test). And statistically modeling the severity of color vision abnormality according to the total number of false scores of the CFM100H test,

(5) using the results, characterized in that the step of modeling the relationship between the total score of the CFM100H test and the degree of color vision abnormality severity,

And (6) expressing numerical values of color vision correction to provide a standard in color display equipment using the relationship between the total number of false scores of the CFM100H test modeled and the degree of color vision abnormality severity.

Therefore, the present invention is to solve the above problems, and the standard value of the blemish score and color vision abnormality of Faransworth-Munsell 100 Hue test, which is one of the representative color vision inspection used to test the presence, type and degree of color vision abnormality The objective is to develop a method for obtaining relationships and figures.

FIG. 1 is a block diagram illustrating a procedure of a method of obtaining a standard value for generating a color vision abnormality for a color display device using a color vision test result according to an exemplary embodiment of the present invention.

As in Figure 1, in order to achieve the object as described above,

Modeling the spectral cone cell response function of the color weakener according to the degree of color vision abnormality (1000);

Step (2000) for verifying the spectral cone cell response function of the modeled color weakness, by simulating the color recognition characteristics of the color weakness using the spectral cone cell response function of the modeled color weakness, and the cap of the CFM100H test. Verifying whether the color weakness's confusion line is consistent with the theory through the color weakness simulation of the color, and through the color weakness simulation of the color of the cap of the CFM100H test, verifying whether the severity of color abnormality increases with increasing severity, and by measuring the cognitive color difference between the color-abbreviated simulated color and the original color through the color weakness simulation for the color of the CFM100H test cap, Measuring the degree of decrease in color discrimination ability as the degree of abnormality increases;

Through clinical trials, the CFM100H test was performed on the color weaker, and the color-corrected CFM100H test (hereinafter referred to as c-CFM100H test) created by calibrating the color of the CFM100H test cap (hereinafter referred to as c-CFM100H test) was performed. statistically modeling the severity of color vision abnormality according to the total scores of the CFM100H test using a total error score (TES) (3000);

Statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test using the total score of the c-CFM100H test made by correcting the color of the CFM100H test cap (4000);

And expressing the severity of color vision abnormality as a numerical value that can be used as a standard by using the relationship between the total number of false scores of the CFM100H test and the degree of color vision abnormality.

(1) spectral cone cell response function modeling step of color weakness according to color vision abnormality

In modeling the spectral cone cell response function of the color weakener according to the degree of color vision abnormality, two important factors are considered. The first element is the shift in peak sensitivity of the cone, and the second element is the shape of the spectral cone function. In general, the degree of movement of the peak response of the cone cells is known to be up to 20 nanometers (nm) [Ref. 3].

는 색각 이상 정도의 수치적 기술에 대한 원추세포의 정점 반응의 이동 정도치를 나타내는 변수이다.The present invention is characterized in that the degree of movement of the cone peak apex reaction has 9 steps at 2 nanometer intervals from 2 nanometers to 18 nanometers. This aims to quantify the degree of severity of color vision as a standard by dividing it into nine steps from 0.1 to 0.9 in 0.1 steps. Figure 2 is a table showing the degree of movement of the cone cone peak response proposed in the present invention. In Figure 2

Is a variable representing the degree of shift of the peak response of the cone to the numerical technique of color vision abnormality.

The present invention is characterized by using an experimentally measured spectroscopic cone cell response function in estimating the shape of the spectroscopic cone cell response function of the color weaker [Refs. 4, 5, 6, for example, Smith and Pokorny. Reaction function measured by. Figure 3 shows the spectral cone cell response function measured by the experiment as described above.

The present invention is to estimate the shape of the spectral cone cell response function according to the color vision abnormality of the color weakness, using the spectral cone cell response function of the color weakness measured in the experiment to estimate the degree of movement of the cone reaction at each wavelength of the visible wavelength Characterized in that. Figure 4 measures the degree of movement of the cone response at each wavelength measured from the spectroscopic cone cell response function of the color weak. Here, the shift value of the L 'cone cell response, which is the abnormal conical cell of the red abbreviation, was measured from the position of the normal M cone cell response, and the shift value of the M' cone cell response, which is the abnormal cone cell of the green abbreviation, was located Was measured from.

The present invention estimates the degree of movement of the cone reaction as shown in the following equation.

도 5는 모델링된 색약자의 분광원추세포반응함수를 나타낸 것이다. 도 5-(가)는적색약자의 분광원추세포반응함수를 나타낸다. 여기서, 점선은 정상적인 S 원추세포의 분광반응함수를 나타내고, 굵은 실선은 정상적인 M 원추세포의 분광반응함수를 나타내고, 가는 실선은 모델링된 L' 원추세포의 분광반응함수를 나타낸다. 도 5-(나)는 녹색약자의 분광원추세포반응함수를 나타낸다. 여기서, 점선은 정상적인 S 원추세포의 분광반응함수를 나타내고, 굵은 실선은 정상적인 L 원추세포의 분광반응함수를 나타내고, 가는 실선은 모델링된 M' 원추세포의 분광반응함수를 나타낸다. 색약의 심각성 정도가 증가할수록 원추세포 반응의 이동 정도가 증가하며, 실험에서 측정된 원추세포 모델과 같이 짧은 파장에서보다 긴 파장에서 이동 정도가 증가하는 것을 관찰할 수 있다. (2) 모델링된 색약자의 분광원추세포반응함수 검증 단계도 6은 모델링된 색약자의 분광원추세포반응함수로부터 색약 정도와 색 구별 관계를 구하는 과정을 도시한 흐름도를 나타낸다. 도 6에서와 같이, 본 발명은 모델링된 색약자의 분광원추세포반응함수로부터 색약 정도와 색 구별 관계를 구하는 과정으로써, 모델링된 색약자의 분광원추세포반응함수를 이용하여 색약자의 색상 인지 특성을 시뮬레이션하여 단계(2100)와, CFM100H 검사의 캡의 색상에 대한 상기 색약 시뮬레이션을 통하여, 색약자의 혼동선이 이론과 일치하는지의 여부를 검증하는 단계(2200)와, CFM100H 검사의 캡의 색상에 대한 상기 색약 시뮬레이션을 통하여, 색약자의 혼동 영역이 색각 이상의 심각성 정도가 증가함에 따라 넓어지는지의 여부를 검증하는 단계(2300)와, CFM100H 검사 캡의 색상에 대한 상기 색약 시뮬레이션을 통하여, 색약자 시뮬레이션된 색상과 원래 색상의 인지적 색상 차이를 측정함으로써, 색각 이상의 정도 증가에 따른 색 구별 능력의 감소 정도를 측정하는 단계(2400)를 포함하는 것을 특징으로 한다.도 7은 모델링된 색약자의 분광원추세포반응함수를 이용하여 색약자의 색상 인지 특성을시뮬레이션하는 단계(2100)를 나타낸다. 색약자의 색상 인지 특성을 시뮬레이션하기 위하여, 색약자의 색 변환 함수(T ^anomalous, 2120)를 이용하여 원래 색상(RGB, 2110)을 LMS 색 공간의 색상(L1M1S1, 2140)으로 변환한다(2130). 변환된 색상(L1M1S1, 2140)을 다시 정상인의 색 변환 함수의(T ^normal) 역 함수([T ^normal]^-1, 2150)를 이용하여 RGB 색 공간의 색상(R1G1B1, 2170)으로 변환함으로써 색약자가 인지하는 색상을 시뮬레이션한다(2160).도 8은 CFM100H 검사의 캡의 색상에 대한 상기 색약 시뮬레이션을 수행한 결과를 x-y 색 공간에 나타낸 것이다. 도 8-(가)는 적색약자의 색약 시뮬레이션 결과를 나타내며, 도 8-(나)는 녹색약자의 색약 시뮬레이션 결과를 나타낸다. 도 8에서, 색약 시뮬레이션된 색상이 표현하는 색 영역(gamut)이 크게 감소하며, 색약의 심각성 정도가 증가할수록 색 영역이 감소함을 관찰할 수 있다.도 9는 CFM100H 검사의 캡의 색상에 대한 상기 색약 시뮬레이션을 수행한 결과를 0.0에서 1.0까지 정규화된 색도(hue) 공간에 나타낸 것이다. 도 9-(가)는 적색약자의 색약 시뮬레이션 결과를 나타낸다. 0.0에서 1.0까지 정규화된 색도 공간에서 적색약자의 혼동선은 0.1882와 0.7417이다. 도 9-(가)에서, 적색약자의 두 혼동선을 중심으로 색상 차이가 크게 감소하며, 색약의 심각성 정도가 증가할수록 혼동선의 변화없이 혼동 영역의 크기가 증가하는 것을 관찰할 수 있다. 도 9-(나)는 녹색약자의 색약 시뮬레이션 결과를 나타낸다. 0.0에서 1.0까지 정규화된 색도 공간에서 녹색약자의 혼동선은 0.1647와 0.7605이다. 도 9-(나)에서, 녹색약자의 두 혼동선을 중심으로 색상 차이가 크게 감소하며, 색약의 심각성 정도가 증가할수록 혼동선의 변화없이 혼동 영역의 크기가 증가하는 것을 관찰할 수 있다.도 10은CFM100H 검사의 캡의 색상과 그것의 색약 시뮬레이션된 색상과의 인지적 색상 차이를 CIE L*a*b* 색 공간에서 측정한 결과를 나타낸다. 도 10-(가)는 적색약자의 색상 차이를 나타내며, 도 10-(나)는 녹색약자의 색상 차이를 나타내며, 도 10-(다)는 적록색약자의 색상 차이를 나타낸다. 도 10에서, 색약의 심각성 정도가 증가할수록 색상 차이가 증가하는 것을 관찰할 수 있다.(3) 실험에 의한 색약 정도에 따른 색약 모델링 검증 단계상기 색약 모델링 결과에 따라 색각 검사를 수행함으로써, 색약 모델을 검증한다. 이를 위하여, 색약자를 대상으로 CFM100H 검사를 수행하도록 하고, CFM100H 검사 캡의 색상을 보정하여 만든 색 보정된 CFM100H 검사(c-CFM100H 검사)를 수행하도록 한다.도 11은 색약자의 색상 인지 능력을 향상시키기 위한 색상 보정 과정을 도시한 것이다. 먼저, 색상을 표현하기 위한 디스플레이 장치(3100)의 분광 특성(spectral characteristics, 3600)을 측정한다. 분광 특성은 분광방사휘도계(spectroradiometer)를 이용하여 측정할 수 있다. 이는 디스플레이 장치의 색상 관리 모듈(3700)에 전달되어 상기 색약 모델을 이용하여 색 보정함수를 생성해낸다. 이때 색약자(3500)로부터 색약 이상 특성(3900)인 색각 이상 종류 및 심각성 정도가 색약 특성 분석 모듈(3800)에 전달된다. 이로써, 색 보정 모듈(3300)이 완성되면, 디스플레이 장치의 색상(3200)이 색 보정 모듈에 전달되어 색 보정된 색상(3400)을 생성해내고, 마지막으로 상기 색약자(3500)에 전달된다. 본 발명은 도 11의 색 보정 원리를 이용하여, 색 보정된 CFM100H 검사(c-CFM100H 검사)를 생성한다. 도 12는 CFM100H 검사와 c-CFM100H 검사를 이용한 임상 실험 절차를 도시한 것이다. 도 12에서, 색약자는 먼저 CFM100H 검사를 수행한다(3210). CFM100H 검사 캡의 원래 색상(3220)은 색약의 심각성 정도에 따라 색 보정된다(3230). 이로써, 9개의 c-CFM100H 검사가 생성된다(3240). c-CFM100H 검사의 색상은 해당 색약 정도에 따라 색 보정된 색상(3250)을 디스플레이 하게 된다. 마지막으로, 각 검사의 총 오점수 결과를 비교하여 색약 정도에 따른 총 오점수 변화를 분석한다(3260).본 발명은 CFM100H 검사의 총 오점수가 색 보정 이후에 감소하였는지를 검사하는 단계를 포함하는 것을 특징으로 한다. 만일 c-CFM100H 검사의 총 오점수가 정상인으로 진단되는 한계 총 오점수(TES_normal)보다 작은지를 통계적으로 검증한다. 본 발명은 색 보정 전/후의 총 오점수 모집단의 평균을 비교함으로써 검증하거나, t-검정을 통하여 보다 신뢰적인 검증을 수행하는 것을 포함하는 것을 특징으로 하나, 검증 방법이 상기 방법에 한정되지는 않는다.도 13은 107안의 색약자를 대상으로 수행한 t-검정 결과를 나타낸다. 도 13-(a)는 미국에서 모집된 60안의 색약자를 대상으로 수행한 t-검정 결과를 나타내며, 도 13-(b)는 한국에서 모집된 47안의 색약자를 대상으로 수행한 t-검정 결과를 나타낸다. 여기서, 모집단의 총 오점수 평균(TES)이 정상인으로 진단되는 한계 총 오점수(TES_normal)보다 통계적으로 작은지 검증하였다. 도 13에서, 통계적 유의성을 판단하는 p-value가 모두 0.05이하로써, 총 오점수 평균(TES)이 정상인으로 진단되는 한계 총 오점수(TES_normal)보다 작다는 증거가 통계적으로 충분함을 알 수 있다.(4) CFM100H 검사의 총 오점수에 따른 색약의 심각성 정도 모델링본 발명은 CFM100H 검사 캡의 색상을 보정하여 만든 c-CFM100H 검사의 총 오점수를 이용하여 CFM100H 검사의 총 오점수에 따른 색각 이상의 심각성 정도를 통계적으로 모델링하는 단계를 포함하는 것을 특징으로 한다.이를 위해, 먼저 상기 시뮬레이션 단계에서 얻었듯이, 색약의 심각성 정도에 따른 색상 인지 능력의 감소가 선형적 함수를 이룬다고 가정한다. 본 발명은 이 가정을 기반으로 CFM100H 검사의 총오점수에 따른 색약의 심각성 정도 모델링하고자 한다. 본 발명은 아래의 수학식에 의해 CFM100H 검사의 총 오점수에 따라 색약의 심각성 정도를 추정한다.

Figure 5 shows the spectral cone cell response function of the modeled color weakness. 5- (A) shows the spectral cone cell response function of the red weak. Here, the dotted line represents the spectral response function of normal S cone cells, the thick solid line represents the spectral response function of normal M cone cells, and the thin solid line represents the spectral response function of modeled L 'cone cells. 5- (b) shows the spectral cone cell response function of the green abbreviation. Here, the dotted line represents the spectral response function of normal S cone cells, the thick solid line represents the spectral response function of normal L cone cells, and the thin solid line represents the spectral response function of modeled M 'cone cells. As the severity of the colorant increases, the degree of migration of the cone response increases, and the degree of migration increases at longer wavelengths than at shorter wavelengths, such as the cone model measured in the experiment. (2) Verification of Spectral Cone Cell Response Function of Modeled Color Abbreviation FIG. 6 is a flowchart illustrating a process of obtaining a color difference and color discrimination relationship from the spectroscopic cone cell response function of a modeled color weakness. As shown in FIG. 6, the present invention is a process for obtaining a color discrimination relationship between the color weakness and the color weakness of the modeled color weakness of the spectroscopic cone cell response function. Step 2100, and through the color weakness simulation of the color of the cap of the CFM100H test, verifying whether the confusion line of the color weakness is consistent with the theory (2200), and the color weakening of the color of the cap of the CFM100H test. Through simulation, verifying whether the confusion area of the color weaker becomes wider as the degree of severity abnormality increases (2300), and the color weakly simulated color and the original color through the color weakness simulation for the color of the CFM100H test cap. By measuring the cognitive color difference of, we measure the decrease in color discrimination ability as the degree of color vision abnormality increases. And in that it comprises a step (2400), characterized in. FIG. 7 by using the spectral response function of the modeled color weakness cones represents the step 2100 to recognize the color of the color weakness simulate characteristics. In order to simulate the color recognition characteristics of the color weaker, the original color (RGB, 2110) is converted into the color of the LMS color space (L1M1S1, 2140) using the color weakness color conversion function ( T ^anomalous , 2120) (2130). Using the normal of the back of the color conversion function, the transformed color ^{(L1M1S1, 2140) (T normal} ) inverse functions ^{([T normal] -1, 2150} ) is converted by the color (R1G1B1, 2170) in the RGB color space, color weakness Simulate the perceived color (2160). FIG. 8 shows the results of the color simulation simulation of the color of the cap of the CFM100H test in the xy color space. 8- (a) shows the color weakness simulation results of the red weak, and FIG. 8- (b) shows the color weakness simulation results of the green weak. In FIG. 8, it can be observed that the color gamut represented by the color simulated color is greatly reduced, and that the color gamut decreases as the severity of the color drug increases. FIG. 9 shows the color of the cap of the CFM100H test. The color simulation results are shown in a hue space normalized from 0.0 to 1.0. 9- (a) shows the color weakness simulation results of the red weak. In the chromaticity space normalized from 0.0 to 1.0, the confusion lines for the red abbreviations are 0.1882 and 0.7417. In FIG. 9- (a), the color difference is greatly reduced around the two confusion lines of the red abbreviation, and as the severity of the color weakness increases, the size of the confusion region increases without changing the confusion lines. 9- (b) shows the color weakness simulation results of the green weak. In the chromaticity space normalized from 0.0 to 1.0, the confusion lines for the green abbreviations are 0.1647 and 0.7605. In FIG. 9- (b), the color difference is greatly reduced around two confusion lines of the green weak, and as the severity of the color weakness increases, the size of the confusion region increases without changing the confusion line. Shows the cognitive color difference between the color of the cap of the CFM100H test and its color weakly simulated color in the CIE L * a * b * color space. 10- (a) shows the color difference of the red abbreviation, FIG. 10- (b) shows the color difference of the green abbreviation, and FIG. 10- (c) shows the color difference of the red abbreviation. In FIG. 10, it can be observed that the color difference increases as the degree of severity of the color drug increases. (3) Color weakness modeling verification step according to color weakness by experiment By performing a color vision test according to the color weakness modeling result, the color weakness model is verified. To this end, the CFM100H test is performed on the color weaker, and the color corrected CFM100H test (c-CFM100H test) made by correcting the color of the CFM100H test cap is performed. It shows a color correction process for. First, the spectral characteristics 3600 of the display apparatus 3100 for expressing color are measured. Spectroscopic properties can be measured using a spectroradiometer. This is transmitted to the color management module 3700 of the display device to generate a color correction function using the color weakness model. At this time, the color weakness kind and severity, which are color weakness characteristics 3900, are transmitted from the color weakness 3500 to the color weakness characteristic analysis module 3800. Thus, when the color correction module 3300 is completed, the color 3200 of the display device is transmitted to the color correction module to generate the color corrected color 3400, and finally, the color weakening 3500 is transmitted. The invention creates a color corrected CFM100H test (c-CFM100H test) using the color correction principle of FIG. Figure 12 illustrates the clinical trial procedure using the CFM100H test and the c-CFM100H test. In FIG. 12, the color weakener first performs a CFM100H test (3210). The original color 3220 of the CFM100H test cap is color corrected 3230 according to the severity of the color weakness. This produces 9 c-CFM100H tests (3240). The color of the c-CFM100H test will display the color corrected color 3250 according to the color intensity. Finally, the total scores of each test are compared to analyze the total score change according to the color intensity (3260). The present invention includes a step of checking whether the total score of the CFM100H test decreased after color correction. It is characterized by. Statistically verify if the total score on the c-CFM100H test is less than the limit total score (TES _normal ) diagnosed as _normal . The present invention includes verifying by comparing the average of the total scores before / after the color correction or performing a more reliable verification through the t-test, but the verification method is not limited to the above method. Fig. 13 shows the results of the t-test performed on 107 color weak subjects. Figure 13- (a) shows the t-test results performed on 60 color weak people recruited from the United States, Figure 13- (b) shows the t-test results performed on 47 color weak people recruited from Korea Indicates. Here, we tested whether the total mean score (TES) of the population was statistically smaller than the limit total mean score (TES _normal ) diagnosed as _normal . In FIG. 13, it can be seen that the p-values for determining statistical significance are all 0.05 or less, so that the evidence that the total score is less than the limit total score (TES _normal ) diagnosed as _normal is statistically sufficient. have. (4) Modeling the severity of colorants according to the total score of the CFM100H test The present invention uses the total score of the c-CFM100H test made by correcting the color of the CFM100H test cap and the severity of color vision abnormality according to the total score of the CFM100H test And statistically modeling the degree. For this purpose, it is assumed that the reduction of the color perception ability according to the severity of the colorant forms a linear function, as first obtained in the simulation step. Based on this assumption, the present invention intends to model the severity of the colorant according to the total score of the CFM100H test. The present invention estimates the severity of the colorant according to the total score of the CFM100H test by the following equation.

다음으로는, c-CFM100H 검사의 총 오점수들 가운데 보다 작은 총 오점수를 나타내는 색약 정도 집합인 dc를 선택한다. 이때, deltaTES은 임상적 에러를 수용하기 위한 값이다. 이 dc들 가운데, 아래의 수학식을 만족하는 값을 해당 색약자의 색약 정도인 d(TES)로 가정한다.

Next, we choose dc, a set of color weaknesses representing the smaller total of the total number of points in the c-CFM100H test. In this case, deltaTES is a value for accommodating clinical error. Among these dcs, a value satisfying the following equation is assumed to be d (TES), which is the color weakness of the corresponding color weakness.

14 is a value showing the estimated degree of color weakness for the total score of the CFM100H test measured in the color weak. The present invention provides a linear regression ( linear regression). 15 shows the linear regression results for the data. The adjusted R ² value, which represents the significance of the linear regression result, was 0.67, which was statistically significant. The amount of color estimating function compared to the modeled total score is as shown in the following equation.

상기 수식은 TES와 색약 정도가 상호 비례한다는 것을 나타나며, 이 비례 관계는 위 수식 이외의 비례 관계를 나타내는 관계를 모두 포함한다.도 16은 상기 모델링된 총 오점수 대비 색약 정도 추정 함수와 상기 색약자의 원추세포반응함수 모델링을 통한 색약자의 색상 인지 시뮬레이션에서의 색상 차이간의 관계를 비교한 것이다. 두 그래프의 선형 관계가 서로 유사함을 관찰할 수 있다. 따라서, 상기 색약자의 원추세포반응함수 모델링이 색약의 정도를 비교적 정확히 표현하였다고 할 수 있다. 상기에서와 같이 추정된 색약 정도는 수치적으로 서술될 수 있다. 위의 사실로 색 표시 장치에서 색약자를 위한 색 보정을 할 때, 색보정을 위한 색약 정도(deficiency degree)를 구하는 방법은 다음과 같다. 우선 기존의 병원에서 측정이 가능한 FM100H를 이용하거나 그것의 컴퓨터로 만든 CFM100H test 를 통하여 TES를 구한다. 그다음 도 16에 TES를 대비하여 색약정도를 구한다. 이 구한 정도로 위에서 언급한 색 보정을 하여 색표시 장치에 디스플레이 하여 준다. [참고문헌 1] S.J. Yang, Y.M. Ro, J. Nam, J.W. Hong, S.Y. Choi, and J.H. Lee, "Improving visual accessibility for color vision deficiency in MPEG-21," ETRI Journal, vol.25, no.3, pp.195-202, June 2004[참고문헌 2] Y.M. Ro and S.J. Yang, "Color Adaptation for Anomalous Trichromats," Int'l Journal of Imaging Systems and Technologies, vol.14, 16-20, 2004[참고문헌 3] D. McIntyre, "Color Blindness: Causes and Effects," Dalton Publishing, 2002 [참조문헌 4] V.C. Smith and J. Pokorny, "Spectral sensitivity of the foveal cone pigments between 400 and 700 nm," Vision Research, vol.15, pp.161-171, 1975[참조문헌 5] J. Pokorny and V.C. Smith, "Evaluation of single-pigment shift model of anomalous trichromacy," J. Opt. Soc. Am. A, vol.67, no.9, pp.1196-1209, 1977[참조문헌 6] J. Pokorny and V.C. Smith, "Derivation of the photopigment absorption spectra in anomalous trichromat,"J. Opt. Soc. Am. A, vol.63, no.2, pp.232-237, 1977[참조문헌 7] P. DeMarco, J. Pokorny, and V.C. Smith, "Full spectrum cone sensitivity functions for X-chromosome-linked anomalous trichromats," J. Opt. Soc. Am. A, vo.9, no.9, pp.1465-1476, 1992

The formula indicates that the TES and the color intensity are mutually proportional, and the proportional relation includes all the relations representing the proportional relationship other than the above formula. This study compares the relationship between color differences in the color perception simulation of color weaknesses through the modeling of cone response responses. You can observe that the linear relationships of the two graphs are similar to each other. Therefore, it can be said that the cone cell response function modeling of the color weakener expresses the degree of color weakness relatively accurately. The estimated color intensity as described above may be described numerically. As a matter of fact, when color correction for color weakness is performed in a color display device, a method of obtaining a degree of deficiency for color correction is as follows. First, find TES using FM100H which can be measured in existing hospital or through CFM100H test made by computer. Then, the degree of color weakness is calculated in comparison with TES in FIG. To this degree, the color correction mentioned above is displayed on the color display device. [Reference 1] SJ Yang, YM Ro, J. Nam, JW Hong, SY Choi, and JH Lee, "Improving visual accessibility for color vision deficiency in MPEG-21," ETRI Journal, vol. 25, no. 3, pp. 195-202, June 2004 [Ref. 2] YM Ro and SJ Yang, "Color Adaptation for Anomalous Trichromats,"Int'l Journal of Imaging Systems and Technologies, vol. 14, 16-20, 2004 [Ref. 3] D. McIntyre, "Color Blindness: Causes and Effects," Dalton Publishing, 2002 [Ref. 4] VC Smith and J. Pokorny, "Spectral sensitivity of the foveal cone pigments between 400 and 700 nm," Vision Research, vol. 15, pp. 161-171, 1975 [Ref. 5] J. Pokorny and VC Smith, "Evaluation of single-pigment shift model of anomalous trichromacy," J. Opt. Soc. Am. A, vol. 67, no. 9, pp. 1196-1209, 1977 [Ref. 6] J. Pokorny and VC Smith, "Derivation of the photopigment absorption spectra in anomalous trichromat," J. Opt. Soc. Am. A, vol. 63, no. 2, pp. 232-237, 1977 [Ref. 7] P. DeMarco, J. Pokorny, and VC Smith, "Full spectrum cone sensitivity functions for X-chromosome-linked anomalous trichromats," J. Opt. Soc. Am. A, vo.9, no.9, pp. 1465-1476, 1992

As described above, the present invention includes modeling the relationship between the blemish score of the CFM100H test and the standard numerical value of the color vision abnormality, which is one of the representative color vision tests used to inspect the presence, type, and degree of color vision abnormality. Thus, color correction for the color weak in the color display device can be enabled.

Claims (23)

The present invention relates to modeling the relationship between the blemish score of the CFM100H test and the standardized numerical value of color vision abnormality. Modeling the spectral cone cell response function of the color weakener according to the degree of color vision abnormality (1000); (2000) verifying the spectral cone cell response function model according to the color intensity by colorant simulation; Experimentally verifying the spectral cone cell response function model according to color intensity using the FM100H or CFM100H test and the c-CFM100H test (3000); Statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test (4000), Color severity abnormality modeling method comprising the step of expressing the severity of color vision abnormality based on a standard value by using the relationship between the total score of the CFM100H test and the degree of color vision abnormality severity (5000) The method of claim 1, In the step of modeling the spectral cone cell response function of the color weakness according to the degree of color vision abnormality, Color vision abnormality degree Color vision abnormality degree modeling method characterized by considering the shift degree of peak sensitivity of the cone and the shape of the spectral cone response function together The method of claim 2, In the step of modeling the degree of shift (peak sensitivity) of the cone cells, The degree of shift of peak sensitivity of cones according to the severity of color

) Is set as shown in the table below.

The method of claim 2, In the step of modeling the shape of the spectral cone response function, Color vision abnormality modeling method characterized by using the spectral cone cell response function measured by the experiment The method of claim 4, wherein In the step of using the spectroscopic cone cell response function measured in the experiment, Color vision abnormality modeling method characterized by estimating the degree of movement of the cone cell response in the wavelength lamda as in the following equation

The method of claim 1, In the step of modeling the spectral cone cell response function of the color weakness according to the degree of color vision abnormality, The spectral cone cell response function of the red drug is modeled as shown in FIG. Color vision abnormality modeling method characterized by modeling the spectral cone cell response function of the green medicine as shown in (b) of FIG. The method of claim 1, In the step of verifying the spectral cone cell response function according to the color weakness through color weakness simulation, Step 2100 by simulating the color recognition characteristics of the color weakness using the spectroscopic cone cell response function of the modeled color weakness, (2200) verifying whether the confusion line of the color weakness matches the theory through the color weakness simulation of the color of the cap of the CFM100H test; (2300) verifying, through the color weakness simulation of the color of the cap of the CFM100H test, whether the confusion area of the color weakened widens with increasing severity of color vision abnormality; By measuring the color weakness simulation of the color of the CFM100H test cap, by measuring the cognitive color difference between the color weakly simulated color and the original color, the step of measuring the reduction in color discrimination ability according to the increase in color vision abnormality (2400) Color vision abnormality modeling method characterized in that The method of claim 7, wherein In step 2100 by simulating the color recognition characteristics of the color weakness using the spectroscopic cone cell response function of the modeled color weakness, In order to simulate the color recognition characteristic of the color weaker, the original color (RGB, 2110) is converted into the color (L1M1S1, 2140) of the LMS color space by using the color weakness color conversion function 2120 (2130). Simulating 2160 the color perceived by the color weaker by converting the converted colors L1M1S1 and 2140 into the colors of the RGB color space R1G1B1 and 2170 using the inverse function 2150 of the color conversion function of the normal person. Color vision abnormality modeling method characterized in that. The method of claim 7, wherein In step 2200, through the color weakness simulation of the color of the cap of the CFM100H test, whether the confusion line of the color weakness is consistent with the theory, Color gamut abnormality modeling, comprising the step of comparing the color gamut of the original cap color and the simulated color by displaying the results of the color weak simulation of the color of the cap of the CFM100H test in the xy color space Way The method of claim 7, wherein The color weakness simulation of the color of the cap of the CFM100H test verifies whether or not the color weakness confusion line is consistent with the theory (2200) and whether the color weakness widens with increasing severity of color vision abnormality. In step 2300 of verifying, Color vision abnormality modeling method comprising the step of comparing the results of the color weakness simulation for the color of the cap of the CFM100H inspection in the hue space (compare) The method of claim 7, wherein In step 2400, the color weakness simulation of the color of the CFM100H test cap is performed to measure the cognitive color difference between the color weakness simulated color and the original color, thereby measuring the degree of reduction of color discrimination ability according to the increase in color vision abnormality. , Colorimetric abnormality modeling method comprising the step of comparing the cognitive color difference between the color of the cap of the CFM100H inspection and its color weakly simulated color in the CIE L * a * b * color space The method of claim 7, wherein In step 2400, the color weakness simulation of the color of the CFM100H test cap is performed to measure the cognitive color difference between the color weakness simulated color and the original color, thereby measuring the degree of reduction of color discrimination ability according to the increase in color vision abnormality. , Colorimetric abnormality modeling method comprising the step of comparing the cognitive color difference between the color of the cap of the CFM100H inspection and its color weakly simulated color in the CIE L * a * b * color space The method of claim 1, In the step (3000) of clinically verifying the spectroscopic cone cell response function according to the color intensity using the CFM100H test and the c-CFM100H test (3000), Color vision abnormality modeling method comprising performing a CFM100H test for the color weak, and performing a color-corrected CFM100H test (c-CFM100H test) made by correcting the color of the CFM100H test cap The method of claim 13, In the step of correcting the color to improve the color recognition ability of the color weak, Measuring the spectral characteristics 3600 of the display apparatus 3100; Spectroscopic characteristics are transmitted to the color management module 3700 of the display device to use the color weakness model, and at the same time, the color weakness type and the severity of the color weakness characteristic 3900 from the color weakness 3500 are the color weakness characteristic analysis module 3800. And color correction 3300 according to the color vision abnormality characteristic of the individual, Transmitting the color 3200 of the display device to the color correction module to generate the color corrected color 3400; Finally, the color vision abnormality modeling method comprising the step of displaying on the color weakening 3500 The method of claim 13, In performing the CFM100H test and the c-CFM100H test, The color weakener performing the CFM100H test (3210); Color correcting the original color 3220 of the CFM100H test cap according to the severity of the colorant (3230), Performing a color-corrected c-CFM100H test for all degrees of color vision anomaly (3240), Lastly, comparing the total scores of the results of each test to analyze the total number of scores according to the degree of color weakness (3260) comprising a color vision abnormality degree modeling method characterized in that it comprises The method of claim 15, In the step (3260) of comparing the total scores of the results of each test to analyze the total score change according to the degree of color weakness, Checking whether the total number of blemishes of the CFM100H test has decreased after color correction. The method of claim 16, In the step of checking whether the total score of the CFM100H test decreased after color correction, Verification by comparing the average of the total number of scores before and after color correction, or performing a more reliable verification through the t-test, color vision abnormality modeling method characterized in that The method of claim 1, In the step 4000 of statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test, Color vision abnormality modeling method characterized by estimating the severity of the colorant according to the total score of the CFM100H test by the following equation

The method of claim 1, In the step 4000 of statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test, A color vision abnormality modeling method characterized by selecting a set of color weaknesses representing a total score less than the sum of TES and clinical error of a normal subject among the total scores of the c-CFM100H test. The method of claim 1, In the step 4000 of statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test, Among the dc, a color vision abnormality modeling method characterized in that to model the value of the color weakness of the color weakness that satisfies the following equation

The method of claim 1, In the step 4000 of statistically modeling the severity of color vision abnormality according to the total score of the CFM100H test, Color regression degree modeling method characterized by using linear regression to accurately model the estimated color intensity function for the total score of the CFM100H test As a method of obtaining a deficiency degree for color correction when performing color correction for a color weakness in a color display device, First, to obtain TES by using FM100H which can be measured in existing hospital or through CFM100H test made by computer Next, the degree of color weakness is calculated for TES in FIG. 16, The color correction is performed to the obtained color weakness and displayed on the color display device. As a device for color correction for color weak in the color display device, First, to obtain TES by using FM100H which can be measured in existing hospital or through CFM100H test made by computer Next, the degree of color weakness is calculated for TES in FIG. 16, A device that corrects the color to about this color and displays it on a color display device.

Images