Color vision ability abnormity detection system based on artificial intelligence

Abstract

The invention discloses a color vision ability abnormity detection system based on artificial intelligence, and particularly relates to the field of color vision science. Comprising a system central processor module, a system operation database, a user information terminal, a photostimulation generation module, a photostimulation data acquisition module, a data preprocessing module, a color vision abnormality comprehensive evaluation and analysis module, a color vision abnormality judgment module and a judgment result output module. According to the invention, through learning and analysis of a large amount of data, an objective judgment standard is established, a deep learning algorithm is utilized to carry out processing on different color combination images, subjective interpretation of a detected person on a pattern is not depended on, and a color vision state is determined through objective judgment of a model according to color features in the images and feedback of the detected person; according to the color vision anomaly detection method and system, the data are rapidly processed through the artificial intelligence system, color vision evaluation of numerous detected persons can be completed in a short time, and the efficiency of color vision anomaly detection is improved.

Description

Technical Field

[0001] The present invention relates to the field of color vision science and technology, and more specifically, to an artificial intelligence-based color vision ability abnormality detection system. Background Art

[0002] Color vision is an important part of the human visual system, which enables humans to perceive and distinguish different colors. Abnormal color vision, such as color blindness and color weakness, is a common visual disorder that affects the human ability to distinguish colors. Color blind people may not be able to accurately identify specific colors such as red, green or blue, while color weak people have a weak ability to distinguish colors. These color vision abnormalities not only affect individuals' daily lives, but may also cause safety hazards in certain occupational fields.

[0003] In the medical and related professional fields, color vision testing is of great significance for diagnosing eye diseases and evaluating occupational adaptability. Traditional color vision testing methods mainly include pseudoisochromatic test, hue arrangement test, color vision mirror test, etc.

[0004] However, there are still some shortcomings in its actual use. For example, traditional color vision detection methods are greatly affected by the subjective understanding and recognition ability of the subject, and therefore cannot accurately determine the degree of color vision abnormality of the subject. At the same time, in scenarios where color vision detection is required for a large number of people, traditional detection methods are inefficient and cannot quickly judge and deal with people with color vision abnormalities. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a color vision ability abnormality detection system based on artificial intelligence, which solves the problems raised in the above-mentioned background technology through the following scheme.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a color vision ability abnormality detection system based on artificial intelligence, comprising a system central processing unit module, a system operation database, a user information terminal, a light stimulation generation module, a light stimulation data acquisition module, a data preprocessing module, a color vision abnormality comprehensive evaluation and analysis module, a color vision abnormality judgment module and a judgment result output module;

[0007] The system operation database includes all data texts of a color vision abnormality detection system based on artificial intelligence, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving a color vision abnormality detection system based on artificial intelligence;

[0008] The light stimulation generation module includes a light stimulation presentation unit and a subject feedback unit, which is responsible for generating various light stimulations for color vision detection and collecting feedback information from the subject after receiving the light stimulation;

[0009] The light stimulation data acquisition module includes a light stimulation parameter measurement unit and a subject perception parameter acquisition unit, which are used to measure various physical parameters of light stimulation during the actual detection process and color-related parameters perceived by the subject, and output the parameters to the data preprocessing module;

[0010] The data preprocessing module is used to clean the collected light stimulation parameters and the perceived parameters of the subject, remove abnormal values ​​and erroneous data, and normalize the data;

[0011] The color vision abnormality comprehensive evaluation and analysis module is used to compare the color vision abnormality comprehensive coefficient value with the color vision abnormality comprehensive coefficient preset value through artificial intelligence, and when the color vision abnormality comprehensive coefficient value is greater than the preset value, it is determined to be color vision abnormality, and the abnormal coefficient f is calculated through artificial intelligence. i (γ, I, S, T) is compared with the abnormal coefficient preset value, and the color with the abnormal coefficient value greater than the abnormal coefficient preset value is screened out to further determine the type of color vision abnormality;

[0012] The color vision abnormality judgment module is used to compare the color vision abnormality comprehensive coefficient value with the color vision abnormality comprehensive coefficient preset value. When the color vision abnormality comprehensive coefficient value is greater than the preset value, it is judged as color vision abnormality, and the abnormal coefficient f is used to determine the color vision abnormality. i (γ, I, S, T) is compared with the abnormal coefficient preset value, and the color with the abnormal coefficient value greater than the abnormal coefficient preset value is screened out to further determine the type of color vision abnormality;

[0013] The judgment result output module is used to import the judgment into the user information terminal.

[0014] Preferably, the light stimulus generating module accurately generates test light of different colors or color combinations according to the set wavelength, light intensity, color saturation and color temperature parameters. For example, it generates test light of a red-green color combination as required, and accurately sets the wavelength of the red light, the wavelength of the green light, and their respective light intensity and color saturation parameters. The module can be equipped with professional optical equipment, such as a light source with adjustable wavelength, a light intensity controller, a color filter, etc., to achieve fine control of various parameters of light stimulation, ensuring that the generated test light meets the detection requirements.

[0015] Preferably, the light stimulation presentation unit is used to adjust the presentation angle and direction of the light stimulation according to the position and sight angle of the subject under a suitable detection environment, and present the light stimulation to the subject in a predetermined color sequence and time interval;

[0016] The subject feedback unit is used to provide a variety of patterns with colors similar to those presented by the light stimulation presentation unit, allowing the subject to select the closest pattern by touch selection, collecting the subject's pattern information and feeding it back to the light stimulation data acquisition module.

[0017] Preferably, the light stimulation parameter measurement unit uses a high-resolution spectrometer to measure the wavelength γ of the light stimulation in real time, accurate to the nanometer level, so as to obtain the wavelength change of the light stimulation during the propagation process, and uses a high-precision photometer to measure the light intensity I at the emission end of the light stimulation and the position of the eyes of the subject, and calculates the attenuation of the light during the propagation process, and uses a professional color analysis instrument to measure the color saturation S and color temperature T of the light stimulation.

[0018] Preferably, the acquisition process of the detected person's perception parameter acquisition unit is as follows:

[0019] A color database is constructed, the database contains n colors, and for each color, its parameters under the same environment are recorded. These parameters are obtained by accurate measurement using a standard color measuring instrument, specifically including RGB value, HSV value, wavelength, light intensity, color temperature and saturation parameters. For example, for the color red, its record in the database is that the RGB value is approximately (255,0,0), the hue in the HSV value is close to 0° or 360°, the saturation is close to 1, and the corresponding wavelength range is 620-760 nanometers. When performing color vision detection, the subject selects a color through the touch screen, and the system obtains the RGB value of the selected color. Subsequently, a matching operation is performed in the color database constructed above to find the color closest to the obtained RGB value. This degree of proximity is determined by measuring the difference between two colors using the Euclidean distance formula in the RGB space, calculating the distance in the RGB space using the Euclidean distance formula, selecting the smallest color as the matching result, and using the wavelength, light intensity, saturation and color temperature corresponding to the matched color as the final subject-perceived color acquisition result for subsequent color vision detection and analysis.

[0020] Preferably, the Euclidean distance formula specifically refers to:

[0021]

[0022] Preferably, the mathematical model of the comprehensive coefficient of color vision abnormality is as follows:

[0023]

[0024] Among them, F(γ,I,S,T) refers to the comprehensive coefficient of color vision abnormality, g i Refers to the weight of the anomaly coefficient of the i-th color, f i(γ,I,S,T) refers to the anomaly coefficient of the i-th color;

[0025] In this technical solution, the g i It is obtained by the entropy method. The specific steps of the entropy method are as follows:

[0026] S1 calculates the proportion of indicators: calculates the proportion Pzj of the zth sample value under the jth indicator to the sum of all sample values ​​of the indicator. This step is the basis for the subsequent calculation of information entropy;

[0027] S2 Calculate information entropy: According to the definition and calculation formula of information entropy, calculate the information entropy ej of the jth indicator. The smaller the information entropy, the greater the degree of variation of the indicator and the greater the amount of information provided;

[0028] S3 calculates information entropy redundancy: information entropy redundancy dj is the complement of information entropy, that is, dj = 1-ej. The larger the information entropy redundancy, the more information the indicator contains, and the greater the impact on the comprehensive evaluation;

[0029] S4 Calculate weight: Calculate the weight of each indicator in the comprehensive evaluation according to its information entropy redundancy. The weight coefficient is the ratio of the information entropy redundancy of each indicator to the sum of the information entropy redundancy of all indicators.

[0030] Preferably, the f i The specific mathematical model of (γ, I, S, T) is as follows:

[0031]

[0032] Among them, γ i Refers to the wavelength of the i-th color collected by the tester, I i Refers to the light intensity of the i-th color collected by the tester, S i Refers to the saturation of the i-th color collected by the tester, T i Refers to the color temperature of the i-th color collected by the tester, Refers to the actual wavelength of the i-th color generated by the light stimulus generation module, Refers to the actual light intensity of the i-th color generated by the light stimulus generation module, Refers to the actual saturation of the i-th color generated by the light stimulus generation module, Refers to the actual color temperature of the i-th color generated by the light stimulus generation module.

[0033] Preferably, the preset value of the color vision abnormality comprehensive coefficient is a critical value of the normal value of the color vision abnormality comprehensive coefficient under the same race, gender, age and environmental conditions obtained according to laboratory practice. When the color vision abnormality comprehensive coefficient is greater than the preset value, it indicates that the subject has color vision abnormality.

[0034] The abnormality coefficient preset value is the critical value of the normal value of the abnormality coefficient of a certain color under the same race, gender, age and environmental conditions obtained based on laboratory practice. When the abnormality coefficient of a certain color is greater than the preset value, it indicates that the person being tested has color vision abnormality for that color.

[0035] Preferably, the user information terminal is an information output device that receives the judgment result of the color vision abnormality judgment module, including a smart phone, a tablet, a computer and other devices that can remotely realize remote communication.

[0036] Technical effects and advantages of the present invention:

[0037] 1. The present invention establishes an objective judgment standard by learning and analyzing a large amount of data, and uses a deep learning algorithm to perform color vision on images of different color combinations. It does not rely on the subjective interpretation of the pattern by the subject, but determines the color vision state based on the characteristics of the color in the image and the feedback of the subject, thereby reducing the detection error caused by individual subjective differences;

[0038] 2. The present invention uses an artificial intelligence system to quickly process data. Through automated testing procedures and rapid algorithm analysis, it can complete color vision assessments for many subjects in a short period of time, thereby improving the efficiency of color vision abnormality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0040] Figure 2 This is a schematic diagram of the structure of the light stimulation generation module of the present invention;

[0041] Figure 3 It is a schematic diagram of the structure of the light stimulation data acquisition module of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] As attached Figure 1 The color vision abnormality detection system shown in the figure is based on artificial intelligence, including a system central processing unit module, a system operation database, a user information terminal, a light stimulation generation module, a light stimulation data acquisition module, a data preprocessing module, a color vision abnormality comprehensive evaluation and analysis module, a color vision abnormality judgment module and a judgment result output module.

[0044] The system operation database includes all data texts of a color vision abnormality detection system based on artificial intelligence, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving a color vision abnormality detection system based on artificial intelligence;

[0045] The output end of the light stimulation generation module is connected to the input end of the light stimulation data acquisition module, the output end of the light stimulation data acquisition module is connected to the input end of the data preprocessing module, the output end of the data preprocessing module is connected to the input end of the color vision abnormality comprehensive evaluation and analysis module, the output end of the color vision abnormality comprehensive evaluation and analysis module is connected to the input end of the color vision abnormality judgment module, and the output end of the color vision abnormality judgment module is connected to the input end of the judgment result output module.

[0046] As attached Figure 2 The light stimulus generation module shown includes a light stimulus presentation unit and a subject feedback unit, which is responsible for generating various light stimuli for color vision detection and collecting feedback information from the subject after receiving the light stimulus;

[0047] What needs to be specifically explained in this embodiment is that the light stimulus generation module accurately generates test light of different colors or color combinations according to the set wavelength, light intensity, color saturation and color temperature parameters. For example, a test light of a red-green color combination is generated according to requirements, and the wavelength of the red light, the wavelength of the green light, and their respective light intensity and color saturation parameters are accurately set. The module can be equipped with professional optical equipment, such as a light source with adjustable wavelength, a light intensity controller, a color filter, etc., to achieve fine control of various parameters of light stimulus, and ensure that the generated test light meets the detection requirements;

[0048] The light stimulation presentation unit is used to adjust the presentation angle and direction of the light stimulation according to the position and sight angle of the subject under a suitable detection environment, and present the light stimulation to the subject in a predetermined color sequence and time interval;

[0049] The subject feedback unit is used to provide a plurality of patterns with colors similar to those presented by the light stimulation presentation unit, allowing the subject to select the closest pattern by touch selection, collecting the subject's pattern information and feeding it back to the light stimulation data acquisition module;

[0050] As attached Figure 3 The light stimulation data acquisition module shown includes a light stimulation parameter measurement unit and a subject perception parameter acquisition unit, which are used to measure various physical parameters of light stimulation during the actual detection process and color-related parameters perceived by the subject, and output the parameters to the data preprocessing module;

[0051] It should be specifically explained in this embodiment that the light stimulus parameter measurement unit uses a high-resolution spectrometer to measure the wavelength γ of the light stimulus in real time, with an accuracy of nanometers, so as to obtain the wavelength change of the light stimulus during the propagation process, and uses a high-precision photometer to measure the light intensity I at the emission end of the light stimulus and the position of the eyes of the subject, and calculates the attenuation of the light during the propagation process, and uses a professional color analysis instrument to measure the color saturation S and color temperature T of the light stimulus;

[0052] The acquisition process of the detected person's perception parameter acquisition unit is as follows:

[0053] Construct a color database, the database contains n colors, and for each color, its parameters under the same environment are recorded. These parameters are obtained by accurate measurement using a standard color measuring instrument, specifically including RGB value, HSV value, wavelength, light intensity, color temperature and saturation parameters. For example, for the color red, its record in the database is that the RGB value is approximately (255,0,0), the hue in the HSV value is close to 0° or 360°, the saturation is close to 1, and the corresponding wavelength range is 620-760 nanometers. When performing color vision detection, the subject selects a color through the touch screen, and the system obtains the RGB value of the selected color. Subsequently, a matching operation is performed in the color database constructed above to find the color closest to the obtained RGB value. This degree of proximity is determined by measuring the difference between two colors using the Euclidean distance formula in the RGB space, calculating the distance in the RGB space using the Euclidean distance formula, selecting the smallest color as the matching result, and using the wavelength, light intensity, saturation and color temperature corresponding to the matching color as the final subject-perceived color acquisition result for subsequent color vision detection and analysis;

[0054] The Euclidean distance formula specifically refers to:

[0055]

[0056] The data preprocessing module is used to clean the collected light stimulation parameters and the perceived parameters of the subject, remove abnormal values ​​and erroneous data, and normalize the data;

[0057] The color vision abnormality comprehensive evaluation and analysis module is used to import the pre-processed data into the mathematical model of the color vision abnormality comprehensive coefficient for processing and analysis, obtain the color vision abnormality comprehensive coefficient, and output the color vision abnormality comprehensive coefficient to the color vision abnormality judgment module;

[0058] It should be specifically explained in this embodiment that the mathematical model of the comprehensive coefficient of color vision abnormality is as follows:

[0059]

[0060] Among them, F(γ,I,S,T) refers to the comprehensive coefficient of color vision abnormality, g i Refers to the weight of the anomaly coefficient of the i-th color, f i (γ,I,S,T) refers to the anomaly coefficient of the i-th color;

[0061] In this technical solution, the g i It is obtained by the entropy method. The specific steps of the entropy method are as follows:

[0062] S1 calculates the proportion of indicators: calculates the proportion Pzj of the zth sample value under the jth indicator to the sum of all sample values ​​of the indicator. This step is the basis for the subsequent calculation of information entropy;

[0063] S2 Calculate information entropy: According to the definition and calculation formula of information entropy, calculate the information entropy ej of the jth indicator. The smaller the information entropy, the greater the degree of variation of the indicator and the greater the amount of information provided;

[0064] S3 calculates information entropy redundancy: information entropy redundancy dj is the complement of information entropy, that is, dj = 1-ej. The larger the information entropy redundancy, the more information the indicator contains, and the greater the impact on the comprehensive evaluation;

[0065] S4 Calculate weight: Calculate the weight of each indicator in the comprehensive evaluation according to its information entropy redundancy. The weight coefficient is the ratio of the information entropy redundancy of each indicator to the sum of the information entropy redundancy of all indicators.

[0066] The f i The specific mathematical model of (γ, I, S, T) is as follows:

[0067]

[0068] Among them, γ i Refers to the wavelength of the i-th color collected by the tester, I i Refers to the light intensity of the i-th color collected by the tester, S i Refers to the saturation of the i-th color collected by the tester, T i Refers to the color temperature of the i-th color collected by the tester, Refers to the actual wavelength of the i-th color generated by the light stimulus generation module, Refers to the actual light intensity of the i-th color generated by the light stimulus generation module, Refers to the actual saturation of the i-th color generated by the light stimulus generation module, Refers to the actual color temperature of the i-th color generated by the light stimulus generation module;

[0069] The color vision abnormality judgment module is used to compare the color vision abnormality comprehensive coefficient value with the color vision abnormality comprehensive coefficient preset value through artificial intelligence, and when the color vision abnormality comprehensive coefficient value is greater than the preset value, it is judged as color vision abnormality, and the abnormal coefficient f is calculated through artificial intelligence. i (γ, I, S, T) is compared with the abnormal coefficient preset value, and the color with the abnormal coefficient value greater than the abnormal coefficient preset value is screened out to further determine the type of color vision abnormality;

[0070] It should be specifically explained in this embodiment that the preset value of the color vision abnormality comprehensive coefficient is the critical value of the normal value of the color vision abnormality comprehensive coefficient under the same race, gender, age and environmental conditions obtained according to laboratory practice. When the color vision abnormality comprehensive coefficient is greater than the preset value, it indicates that the subject has color vision abnormality;

[0071] The abnormal coefficient preset value is the critical value of the normal value of the abnormal coefficient of a certain color under the same race, gender, age and environmental conditions obtained according to laboratory practice. When the abnormal coefficient of a certain color is greater than the preset value, it indicates that the subject has color vision abnormality for the color;

[0072] The judgment result output module is used to import the judgment into the user information terminal;

[0073] It should be specifically explained in this embodiment that the user information terminal is an information output device that receives the judgment result of the color vision abnormality judgment module, including a smart phone, a tablet, a computer and other devices that can remotely realize remote communication;

[0074] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0075] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A color vision abnormality detection system based on artificial intelligence, characterized in that: include: System central processing unit module, system operation database, user information terminal, light stimulation generation module, light stimulation data acquisition module, data preprocessing module, color vision abnormality comprehensive evaluation and analysis module, color vision abnormality judgment module and judgment result output module; The system operation database includes all data texts of a color vision abnormality detection system based on artificial intelligence, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving a color vision abnormality detection system based on artificial intelligence; The light stimulation generation module includes a light stimulation presentation unit and a subject feedback unit, which is responsible for generating various light stimulations for color vision detection and collecting feedback information from the subject after receiving the light stimulation; The light stimulation data acquisition module includes a light stimulation parameter measurement unit and a subject perception parameter acquisition unit, which are used to measure various physical parameters of light stimulation during the actual detection process and color-related parameters perceived by the subject, and output the parameters to the data preprocessing module; The data preprocessing module is used to clean the collected light stimulation parameters and the perceived parameters of the subject, remove abnormal values ​​and erroneous data, and normalize the data; The color vision abnormality comprehensive evaluation and analysis module is used to import the pre-processed data into the mathematical model of the color vision abnormality comprehensive coefficient for processing and analysis, obtain the color vision abnormality comprehensive coefficient, and output the color vision abnormality comprehensive coefficient to the color vision abnormality judgment module; The color vision abnormality judgment module is used to compare the color vision abnormality comprehensive coefficient value with the color vision abnormality comprehensive coefficient preset value through artificial intelligence, and when the color vision abnormality comprehensive coefficient value is greater than the preset value, it is judged as color vision abnormality, and the abnormal coefficient f is calculated through artificial intelligence. i (γ, I, S, T) is compared with the abnormal coefficient preset value, and the color with the abnormal coefficient value greater than the abnormal coefficient preset value is screened out to further determine the type of color vision abnormality; The judgment result output module is used to import the judgment into the user information terminal.

2. The color vision abnormality detection system based on artificial intelligence according to claim 1, characterized in that: The light stimulation presentation unit is used to adjust the presentation angle and direction of the light stimulation according to the position and sight angle of the subject under a suitable detection environment, and present the light stimulation to the subject in a predetermined color sequence and time interval; The subject feedback unit is used to provide a variety of patterns with colors similar to those presented by the light stimulation presentation unit, allowing the subject to select the closest pattern by touch selection, collecting the subject's pattern information and feeding it back to the light stimulation data acquisition module.

3. The color vision abnormality detection system based on artificial intelligence according to claim 1, characterized in that: The light stimulation parameter measurement unit uses a high-resolution spectrometer to measure the wavelength γ of the light stimulation in real time, uses a high-precision photometer to measure the light intensity I at the emission end of the light stimulation and the position of the eyes of the subject, and calculates the attenuation of the light during the propagation process, and uses a professional color analysis instrument to measure the color saturation S and color temperature T of the light stimulation.

4. The color vision abnormality detection system based on artificial intelligence according to claim 1, characterized in that: The acquisition process of the detected person's perception parameter acquisition unit is as follows: A color database is constructed, which contains n colors, and for each color, its parameters under the same environment are recorded. These parameters are obtained by accurate measurement through standard color measuring instruments, specifically including RGB value, HSV value, wavelength, light intensity, color temperature and saturation parameters. When performing color vision detection, the subject selects a color through the touch screen, and the system will obtain the RGB value of the selected color. Subsequently, a matching operation is performed in the color database constructed above to find the color closest to the obtained RGB value, and the distance in the RGB space is calculated by the Euclidean distance formula. The smallest color is selected as the matching result, and the wavelength, light intensity, saturation and color temperature corresponding to the matched color are used as the final subject-perceived color acquisition result for subsequent color vision detection and analysis.

5. The color vision abnormality detection system based on artificial intelligence according to claim 1, characterized in that: The mathematical model of the color vision abnormality comprehensive coefficient is as follows: Among them, F(γ,I,S,T) refers to the comprehensive coefficient of color vision abnormality, g i Refers to the weight of the anomaly coefficient of the i-th color, f i (γ,I,S,T) refers to the anomaly coefficient of the i-th color.

6. The color vision abnormality detection system based on artificial intelligence according to claim 1, characterized in that: The f i The specific mathematical model of (γ, I, S, T) is as follows: Among them, γ i Refers to the wavelength of the i-th color collected by the tester, I i Refers to the light intensity of the i-th color collected by the tester, S i Refers to the saturation of the i-th color collected by the tester, T i Refers to the color temperature of the i-th color collected by the tester, Refers to the actual wavelength of the i-th color generated by the light stimulus generation module, Refers to the actual light intensity of the i-th color generated by the light stimulus generation module, Refers to the actual saturation of the i-th color generated by the light stimulus generation module, Refers to the actual color temperature of the i-th color generated by the light stimulus generation module.

Images