Intelligent eye-protection illumination control system

Abstract

The invention discloses an intelligent eye-protection illumination control system, and relates to the technical field of intelligent illumination control, and the system comprises a physiological analysis module which is used for calculating the pupil diameter physiological index of a user through a Daugman algorithm based on the preprocessed multi-source data, marking fatigue early warning, obtaining the fatigue state of the user, and obtaining the fatigue state of the user according to the long-time fixed region of the sight line of the user; performing sight line drop point processing and region division through a space coordinate mapping and clustering algorithm to generate a concentration region coordinate; the instruction generation module is used for fusing the scene initial light strategy and the focus area coordinate through a weighted fusion method, forming a light parameter in combination with a user vision parameter, and analyzing through an NSGA-II algorithm to obtain an illumination control instruction; the execution module is used for transmitting and executing based on the illumination control instruction through a wireless communication method, and monitoring visual physiological reaction of the user in real time to obtain feedback data; according to the invention, the environment change is predicted through the Markov model, the illumination is prevented from lagging behind the environment change, and the illumination foresight and adaptability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and particularly to an intelligent eye protection lighting control system. Background Art

[0002] With the growing demand for healthy lighting, intelligent lighting technology has developed rapidly. Traditional intelligent lighting systems are mostly based on single or a small number of parameters such as ambient light intensity and human body infrared induction, and control the on / off and brightness adjustment of lights through preset programs, and have gradually developed into systems that can adjust light parameters according to simple scene modes (such as reading and rest modes). These technologies have met the basic needs of users for lighting intelligence to a certain extent and promoted the transformation of the lighting industry from traditional to intelligent.

[0003] Currently, there is a common problem of insufficient personalized adaptation in intelligent lighting systems, and it is difficult to dynamically optimize the light environment according to physiological parameters such as users' visual characteristics, light sensitivity, and color vision differences. Traditional solutions only adjust based on ambient light or simple preset modes, and cannot accurately control in combination with users' real-time visual fatigue status, work scene requirements (such as reading and screen use), and eye health data (such as diopter and glare sensitivity), resulting in a mismatch between lighting parameters and individual eye protection needs. In the long term, it may exacerbate problems such as visual fatigue and dry eyes, affecting visual comfort and health protection effects. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent eye protection lighting control system to solve the problem that the lighting light environment is difficult to adapt to personalized eye protection needs.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: The present invention provides an intelligent eye - protection lighting control system, which includes a data acquisition module that collects multi - source data in real - time, pre - processes it, and analyzes it through a Markov model to obtain an initial light strategy for the scene; a physiological analysis module that calculates the physiological index of the user's pupil diameter and marks fatigue warnings based on the pre - processed multi - source data through the Daugman algorithm to obtain the user's fatigue state, and performs gaze landing point processing and area division through spatial coordinate mapping and clustering algorithms according to the area where the user's line of sight is fixed for a long time to generate the coordinates of the focused area; an instruction generation module that fuses the initial light strategy of the scene and the coordinates of the focused area through a weighted fusion method, combines the user's vision parameters to form light parameters, and analyzes them through the NSGA - II algorithm to obtain lighting control instructions; an execution module that transmits and executes based on the lighting control instructions through a wireless communication method, and monitors the user's visual physiological responses in real - time to obtain feedback data; an optimization module that dynamically corrects the light parameters using a fuzzy logic algorithm based on the user's fatigue state to obtain optimized light parameters, and generates optimal intelligent eye - protection lighting control parameters through data fusion and intelligent decision - making algorithms in combination with the feedback data.

[0007] As a preferred solution of the intelligent eye - protection lighting control system described in the present invention, wherein: the multi - source data includes environmental light intensity data, environmental color temperature data, user pupil diameter data, user blink frequency data, and data on the area where the user's line of sight is fixed for a long time; The pre - processing includes standardization, noise removal, filling missing values, and outlier processing.

[0008] As a preferred solution of the intelligent eye - protection lighting control system described in the present invention, wherein: the process of analyzing through the Markov model to obtain the initial light strategy for the scene is as follows: Based on the historical multi - source data after pre - processing, train the Markov model through the maximum likelihood estimation method to obtain the trained Markov model; Input the pre - processed multi - source data into the Markov model, and predict the change trend of environmental light intensity in the current scene according to the transition probability relationship between different environmental and user state data, and analyze the corresponding preliminary light strategy parameters for different scenes; Sort and screen the preliminary light strategy parameters through the analytic hierarchy process to obtain the initial light strategy for the scene.

[0009] As a preferred solution of the intelligent eye - protection lighting control system described in the present invention, wherein: the process of calculating the user's physiological index and marking fatigue warnings based on the pre - processed multi - source data through the Daugman algorithm to obtain the user's fatigue state is as follows: Input the pre - processed multi - source data into the Daugman algorithm, extract the pupil and iris boundaries through circle detection and bilinear interpolation, and calculate the physiological index of the change rate of the user's pupil diameter; Based on historical fatigue data and historical normal state data, parameter optimization is carried out by means of the support vector machine method to obtain a fatigue threshold; The physiological index of the user's pupil diameter change rate and the user's blink frequency data are compared with the fatigue threshold to obtain the user's fatigue state.

[0010] As a preferred scheme of the intelligent eye protection lighting control system described in the present invention, wherein: according to the area where the user's line of sight is fixed for a long time, through spatial coordinate mapping and clustering algorithms, the line of sight landing point is processed and the area is divided to generate the coordinates of the focused area. The specific steps are as follows. The area where the user's line of sight is fixed for a long time is converted into a unified standardized coordinate system through the spatial coordinate mapping algorithm; Through the DBSCAN clustering algorithm, clustering is carried out according to the spatial position similarity of the area where the user's line of sight is fixed for a long time, the preliminary focused area is divided, and abnormal landing points are removed and the boundary is smoothed; Based on the position of the preliminary focused area in the standardized coordinate system, analysis is carried out through the minimum bounding rectangle method to obtain the coordinates of the focused area.

[0011] As a preferred scheme of the intelligent eye protection lighting control system described in the present invention, wherein: the initial light strategy of the scene and the coordinates of the focused area are fused by a weighted fusion method, and combined with the user's vision parameters to form light parameters. The specific steps are as follows. Based on the initial light strategy data of the scene and the coordinates of the focused area, the influence degree and correlation of the user's visual experience are analyzed by the principal component analysis method to obtain the influence weight of the user's visual experience, and the weighted fusion method is used for fusion to obtain the light strategy; The light strategy is combined with the user's vision parameters, and through an adaptive adjustment algorithm, optimization and matching are carried out to form light parameters suitable for the user.

[0012] As a preferred scheme of the intelligent eye protection lighting control system described in the present invention, wherein: the lighting control instruction is obtained through the NSGA-II algorithm. The specific steps are as follows. Based on the light parameters, multi-objective functions of lighting energy conservation and visual comfort are set, and a population of parameter combinations of lighting control instructions containing multiple individuals is randomly generated and initialized; Non-dominated sorting and crowding degree analysis are carried out on the population of parameter combinations of lighting control instructions, through selection, crossover, mutation, merging, updating and elite retention; When the multi-objective functions of lighting energy conservation and visual comfort converge, the optimal individual is selected from the population of parameter combinations of lighting control instructions to obtain the lighting control instruction.

[0013] As a preferred embodiment of the intelligent eye - care lighting control system of the present invention, it is as follows: The lighting control instruction is transmitted and executed through a wireless communication method, and the visual physiological response of the user is monitored in real - time to obtain feedback data. The specific steps are as follows. Encode the lighting control instruction, convert it into the JSON format for wireless communication transmission, and send it to the lighting end through the wireless communication protocol. The lighting end receives the lighting control instruction transmitted wirelessly, decodes and restores it to the original lighting control instruction, and adjusts and executes the lighting parameters according to the lighting control instruction. Monitor the visual physiological response data of the user in real - time and perform pre - processing to extract the feedback data related to the visual physiological response of the user.

[0014] As a preferred embodiment of the intelligent eye - care lighting control system of the present invention, it is as follows: The light parameters are dynamically corrected using a fuzzy logic algorithm based on the user's fatigue state to obtain optimized light parameters. The specific steps are as follows. Fuzzify the user's fatigue state through a fuzzy logic algorithm, convert it into a fuzzy linguistic variable, and perform logical reasoning through a fuzzy inference method in combination with the fuzzy rule base to obtain a fuzzy conclusion for light parameter adjustment. Use the centroid method to defuzzify the fuzzy conclusion and convert it into a numerical value to obtain the optimized light parameters.

[0015] As a preferred embodiment of the intelligent eye - care lighting control system of the present invention, it is as follows: Combine the optimized light parameters with the feedback data, perform integration processing through weighted average fusion, and conduct in - depth analysis using Bayesian decision - making to generate the optimal intelligent eye - care lighting control parameters.

[0016] The beneficial effects of the present invention are as follows: By predicting environmental changes through a Markov model, it can pre - sense the evolution trend of factors such as environmental light intensity in advance, pre - adjust the light strategy, and avoid lighting lagging behind environmental changes; it improves the lighting's forward - looking and adaptability, reduces the impact of environmental mutations on lighting effects, creates a more stable and comfortable lighting environment for users, and enhances the user experience; the fuzzy logic algorithm responds to the fatigue state, can convert the complex and fuzzy fatigue state of users into precise light parameter adjustment instructions; it can flexibly adapt to the lighting environment according to different fatigue degrees, effectively relieve visual fatigue, provide a personalized and dynamic eye - care lighting solution for users, and improve the lighting intelligence and user comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic diagram of an intelligent eye protection lighting control system.

[0019] Figure 2 It is a schematic diagram of the generation of the initial light strategy for the scene.

[0020] Figure 3 It is a schematic diagram of the physiological analysis of the user.

[0021] Figure 4 It is a schematic diagram of the generation of intelligent eye protection lighting control parameters. Specific Embodiments

[0022] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings of the specification.

[0023] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0025] Refer to Figures 1 to 4 , which is an embodiment of the present invention. This embodiment provides an intelligent eye protection lighting control system, including the following steps: A data acquisition module that collects multi-source data in real time for preprocessing.

[0026] The multi-source data includes ambient light intensity data, ambient color temperature data, user pupil diameter data, user blink frequency data, and user long-term fixed gaze area data.

[0027] Specifically, the ambient light intensity data, which uses a light intensity sensor to collect the ambient light intensity values ​​of the working area in front of the user, the area around the screen, the area facing the light source, and the area around the user's head at different time points; the ambient color temperature data, which uses a color temperature sensor to analyze the color temperature of the light around the user and extract the color temperature value; the user's pupil diameter data, which uses a camera to collect the user's eye image to obtain the pupil diameter parameter; the user's blinking frequency data, which detects the user's eyelid opening and closing status through a continuous image frame sequence, and counts the number of blinks through a time window to obtain the user's blinking frequency per unit time; the user's line of sight long-term fixed area data, which uses a camera to track the user's eye movement trajectory, and accumulates the coordinates of the line of sight landing point over time to form the user's line of sight long-term fixed area data.

[0028] Preprocessing includes standardization, noise removal, missing value filling and outlier processing.

[0029] Specifically, the mean and standard deviation of the ambient light intensity data, ambient color temperature data, user pupil diameter data, user blinking frequency data and user's long-term fixed line of sight data are calculated respectively by the Z-score normalization method. The corresponding mean is subtracted from each ambient light intensity data, ambient color temperature data, user pupil diameter data, user blinking frequency data and user's long-term fixed line of sight data point and then divided by the standard deviation to complete the Z-score normalization method standardization processing; for the ambient light intensity data, ambient color temperature data, user pupil diameter data and user blinking frequency data, a fixed-length sliding window containing the current point is selected at each time point, and the ambient light intensity data, ambient color temperature data, user pupil diameter data and user blinking frequency data in the fixed-length sliding window are sorted from small to large, and then the ambient light intensity data, ambient color temperature data, user pupil diameter data and user blinking frequency data in the middle position are taken. The data is used as the new value at the current time point to complete the noise removal using the median filtering method; for all missing positions marked as empty in the user pupil diameter data and the user blinking frequency data, the Euclidean distance between the missing position and the non-missing pupil diameter data and the user blinking frequency data point is analyzed, the non-missing position with the smallest distance is selected, and the non-missing position is weighted averaged as the filling point for the missing position to complete the missing filling processing using the K-nearest neighbor interpolation method; for the ambient light intensity data, ambient color temperature data, user pupil diameter data, and user blinking frequency data, the upper quartile and the lower quartile are calculated respectively, and then the interquartile range is calculated, and the upper boundary of the outlier is the upper quartile plus one times the interquartile range, and the lower boundary of the outlier is the lower quartile minus one times the interquartile range, the values ​​beyond the boundary are marked as outliers and are removed from the ambient light intensity data, ambient color temperature data, user pupil diameter data, and user blinking frequency data to complete the outlier processing using the box plot method.

[0030] The initial light strategy of the scene is obtained through analysis of the Markov model.

[0031] Based on the pre - processed historical multi - source data, the Markov model is trained by the maximum likelihood estimation method to obtain the trained Markov model.

[0032] Specifically, the environmental light intensity data, environmental color temperature data, user pupil diameter data, user blink frequency data, and user long - term fixed gaze area data are sequentially extracted from the pre - processed historical multi - source data. Each group of time - synchronized environmental light intensity data is divided into four exemplary intervals: 0 to 200 lux, 200 to 500 lux, 500 to 800 lux, and above 800 lux. The environmental color temperature data is divided into four exemplary intervals: 2500 to 3500 Kelvin, 3500 to 5000 Kelvin, 5000 to 6500 Kelvin, and above 6500 Kelvin. The user pupil diameter data is divided into four exemplary intervals: 2 to 3 mm, 3 to 4 mm, 4 to 5 mm, and above 5 mm. The interval numbers of each group of corresponding positions in the above three intervals are combined to form state labels, forming a state label sequence arranged in chronological order. All state transition pairs composed of two adjacent state labels in the state label sequence are enumerated, the frequency of each type of state transition pair is counted, the sum of the frequencies of transitioning from each state label to all other state labels is recorded, and the frequency of transitioning from each state label to another state label is divided by the sum of all transition frequencies of the state label to obtain the transition probability between state labels. All the transition probability values between state labels are organized to form a state transition probability matrix. According to the maximum likelihood estimation method, all probability values in the state transition probability matrix are subjected to a joint probability multiplication operation to evaluate the fitting degree of the state transition probability matrix with the observed data, thereby verifying the rationality of the Markov model. By dividing the frequency of each state transition pair in the state label sequence by the total transition frequency of the corresponding starting state label and directly assigning it to the probability value at the corresponding position in the state transition probability matrix, the joint probability product corresponding to it in the state label sequence reaches the maximum value to obtain the trained Markov model.

[0033] The pre - processed multi - source data is input into the Markov model. Based on the transition probability relationship between different environmental and user state data, the change trend of the environmental light intensity in the current scenario is predicted, and the corresponding preliminary light strategy parameters in different scenarios are analyzed.

[0034] Specifically, the environmental light intensity, environmental color temperature, user pupil diameter, blink frequency, and gaze - fixed area data collected in real - time are respectively divided into intervals, and the interval numbers corresponding to the above five types of data at the current moment are extracted in chronological order and spliced to form the current state label; The construction of the optical strategy parameter configuration rule table is based on environmental light intensity data, environmental color temperature data, user pupil diameter data, user blink frequency data, and user long-term fixed gaze area data, which are divided into multiple intervals. Each combination represents a specific user state and environmental condition. By analyzing historical operation data and experimental test results, a correlation model between different state combinations and indicators such as user visual comfort and work efficiency is established; then a multi-objective optimization algorithm is adopted to match the light intensity and color temperature parameters that can bring the best visual experience for each state combination while meeting the energy-saving requirements; finally, these verified parameter configurations are filled into the corresponding positions in the rule table to form a complete "state-lighting strategy" mapping relationship, ensuring that there is a corresponding lighting setting for each state. With the collection and analysis of new environmental light intensity data, environmental color temperature data, user pupil diameter data, user blink frequency data, and user long-term fixed gaze area data, the optical strategy parameter configuration rule table will be continuously updated to adapt to different environmental changes and user needs; Input the current state label into the Markov model, obtain all target state labels and transition probability values corresponding to the current state label in the state transition probability matrix. Based on the state transition probability of the Markov model, perform weighted analysis on all possible target states, comprehensively consider the typical lighting characteristics and their transition possibilities of each state, and finally calculate the expected change direction of the light intensity to obtain the expected trend value of the environmental light intensity change. Divide the expected trend value into four exemplary light intensity levels: 100 to 300 lux, 300 to 500 lux, 500 to 700 lux, and above 700 lux. Combine the current environmental color temperature data interval number, user pupil diameter data interval number, and light intensity level to form a triple index, and look up the light strategy parameter setting value matched by the corresponding triple from the optical strategy parameter configuration rule table to output the preliminary light strategy parameters corresponding to different scenarios.

[0035] Sort and screen the preliminary light strategy parameters through the analytic hierarchy process to obtain the initial light strategy for the scenario.

[0036] Specifically, through the analytic hierarchy process, a hierarchical structure of decision problems for the initial light strategy of the scene is generated, and multiple factors affecting the lighting parameters (ambient light intensity data, ambient color temperature data, user pupil diameter data, user blink frequency data, and user long-term fixed gaze area data) are stratified according to their importance in lighting control; each level contains specific decision criteria. For example, in the ambient level, the changes in ambient light intensity and color temperature are considered, and in the user physiological state level, data such as user pupil diameter and blink frequency are considered. Based on the above criteria, by pairwise comparing multiple factors affecting the lighting parameters, the relative importance of multiple factors affecting the lighting parameters is determined. Usually, the scale method is used for evaluation, and the comparison results of multiple factors affecting the lighting parameters are expressed numerically to form a judgment matrix; through matrix operations and consistency tests, the weight priorities of each factor are determined by solving the eigenvalues and eigenvectors of the judgment matrix to obtain the weight values of the factors of each lighting parameter. According to the weight values of the factors of each lighting parameter, the most important factors of the lighting parameters are sorted, and the priority of each initial light strategy parameter of the scene is determined using the weight values of the factors of the lighting parameters. Through this method, the most suitable preliminary light strategies under different environments and user states are screened out.

[0037] The physiological analysis module calculates user physiological indicators and marks fatigue warnings through the Daugman algorithm based on the preprocessed multi-source data to obtain the user's fatigue state.

[0038] The preprocessed multi-source data is input into the Daugman algorithm, and through circle detection and bilinear interpolation, the pupil and iris boundaries are extracted to calculate the physiological indicator of the user's pupil diameter change rate.

[0039] It should be noted that the expression for calculating the physiological indicator of the user's pupil diameter change rate is: ; where R is the physiological indicator of the pupil diameter change rate, D t1 is the user pupil diameter data at the first moment t1, and D t2 is the user pupil diameter data at the second moment t2; Specifically, the preprocessed multi-source data is input into the Daugman algorithm, and the circle detection method is used to locate the candidate center and radius of the pupil and iris boundaries in the user's pupil diameter data image. The integral differential operator traverses the user's pupil diameter data image space. The integral differential operator traverses the pixel points of the user's pupil diameter data image along the circular path generated by the candidate center and radius, and performs an accumulation operation on the change amplitude of the pixel gray values of the user's pupil diameter data image on the circular path. The accumulation result is used as the edge intensity value. The integral differential operator moves along the circular path through each pixel of the user's pupil diameter data image in the user's pupil diameter data image space with the center of the circle as the center and the candidate radius length as the distance, extracts the gray value difference between adjacent pixels of the user's pupil diameter data image on the path, and sums the absolute values of the differences to generate the edge intensity value. The operation of summing the absolute values of the gray value differences is repeatedly performed on the circular paths corresponding to different combinations of the center and radius. The larger the edge intensity value, the closer the current combination of the center and radius is to the true pupil or iris boundary. Identify the geometric parameters of the pupil and iris boundaries corresponding to the maximum edge response value, and output the accurate pupil position and size data; The bilinear interpolation method performs interpolation operations on the pupil and iris boundaries, generates new pixel values based on the weighted average of the gray values of the surrounding four known pixel points, improves the boundary positioning accuracy to the sub-pixel level, extracts the closed curve coordinates of the pupil and iris boundaries from the interpolation results, and fits the boundary curve equation by the least squares method to eliminate noise interference and smooth the geometric shape; calculate the physiological index of the user's pupil diameter change rate and generate the physiological index of the user's pupil diameter change rate.

[0040] Based on the historical fatigue data and historical normal state data, parameter optimization is carried out through the support vector machine method to obtain the fatigue threshold.

[0041] Specifically, the physiological index of the user's pupil diameter change rate and the user's blink frequency data in the historical fatigue data and historical normal state data are used as features to input into the support vector machine method. The historical fatigue data and historical normal state data include the mean value of the physiological index of the user's pupil diameter change rate and the variance of the user's blink frequency data. The support vector machine method uses the radial basis kernel function to map the feature vectors of the historical fatigue data and historical normal state data into a high-dimensional space. During the training process, the grid search method is applied to traverse the combinations of candidate kernel function parameters and penalty coefficients, and the cross-validation method is performed for each candidate combination to evaluate the classification accuracy. The cross-validation method divides the historical fatigue data and historical normal state data into a training set and a validation set. The training set determines the support vectors by solving the Lagrangian dual problem and generates a hyperplane with the maximum classification margin. The parameters of the hyperplane equation are derived from the linear combination of the support vectors and the label values. The validation set tests the generalization performance of the hyperplane. The classification decision function is generated based on the optimal kernel function parameters and penalty coefficients. The optimal parameters correspond to the candidate combination with the highest classification accuracy. After locking, the fatigue threshold is determined by the geometric distance between the classification hyperplane and the origin.

[0042] Compare the physiological index of the user's pupil diameter change rate and the user's blink frequency data with the fatigue threshold to obtain the user's fatigue state.

[0043] Specifically, numerically compare the value of the physiological index of the user's pupil diameter change rate with the fatigue threshold. When the value of the physiological index of the user's pupil diameter change rate exceeds the fatigue threshold, the user's fatigue state is marked as fatigued. When the value of the physiological index of the user's pupil diameter change rate is lower than the fatigue threshold, the user's fatigue state is marked as non-fatigued. Numerically compare the value of the user's blink frequency data collected in real time with the fatigue threshold. When the value of the user's blink frequency data exceeds the fatigue threshold, the user's fatigue state is marked as non-fatigued. When the value of the user's blink frequency data is lower than the fatigue threshold, the user's fatigue state is marked as fatigued. When the comparison result of the physiological index of the user's pupil diameter change rate is marked as fatigued or the comparison result of the user's blink frequency data is marked as fatigued, the user's fatigue state is determined to be fatigued. When the comparison result of the physiological index of the user's pupil diameter change rate is marked as non-fatigued and the comparison result of the user's blink frequency data is marked as non-fatigued, the user's fatigue state is determined to be non-fatigued.

[0044] According to the area where the user's line of sight is fixed for a long time, perform line-of-sight landing point processing and area division through spatial coordinate mapping and clustering algorithms to generate the coordinates of the focused area.

[0045] Convert the area where the user's line of sight is fixed for a long time into a unified standardized coordinate system through a spatial coordinate mapping algorithm.

[0046] Specifically, the original coordinate data in the data of the area where the user's line of sight is fixed for a long time is input into the spatial coordinate mapping algorithm. The original coordinate data includes the horizontal pixel position and the vertical pixel position. The spatial coordinate mapping algorithm determines the horizontal maximum value and the vertical maximum value of the standardized coordinate system based on the screen resolution parameters. The screen resolution parameters include the total number of horizontal pixels and the total number of vertical pixels. The original coordinate data is normalized through the screen resolution parameters, and the horizontal and vertical pixel positions are respectively converted into the standardized coordinate system within the range of 0 to 1. The standardized horizontal coordinate and the standardized vertical coordinate are combined into the coordinate value in the standardized coordinate system. The horizontal coordinate range of the standardized coordinate system is limited to 0 to 1, and the vertical coordinate range is limited to 0 to 1. Each original coordinate point of the data of the area where the user's line of sight is fixed for a long time performs horizontal and vertical standardization conversion operations to generate the standardized coordinate system.

[0047] Through the DBSCAN clustering algorithm, cluster according to the spatial position similarity of the area where the user's line of sight is fixed for a long time, divide out the preliminary focused area, and remove abnormal landing points and smooth the boundary.

[0048] Specifically, the coordinate set of the data in the area where the user's line of sight is fixed for a long time is input into the DBSCAN clustering algorithm in the standardized coordinate system. The DBSCAN clustering algorithm defines the neighborhood radius parameter and the minimum neighborhood sample number parameter. Based on the Euclidean distance formula, it traverses each focus area coordinate point in the focus area coordinate set, counts the number of adjacent coordinate points within the neighborhood radius of each focus area coordinate point, and marks the focus area coordinate points whose number of adjacent focus area coordinate points exceeds the minimum neighborhood sample number parameter as core points. The core points and other coordinate points within the neighborhood are merged into the same clustering cluster. The focus area coordinate points that cannot be assigned to any clustering cluster are marked as noise points, and the noise points are removed from the focus area coordinate set. The remaining coordinate points form the preliminary focus area coordinate points. The preliminary focus area coordinate point set is input into the convex hull generation algorithm. The convex hull generation algorithm selects the preliminary focus area coordinate point with the smallest y coordinate value in the preliminary focus area coordinate point set as the initial extreme point, measures the polar angles of other coordinate points in the coordinate point set relative to the initial extreme point, sorts the preliminary focus area coordinate points in ascending order of the polar angles, and the convex hull generation algorithm sequentially connects adjacent coordinate points in the order of the sorted preliminary focus area coordinate points to form candidate edges. The candidate edges are checked for convexity and concavity with the polygon edges formed by the connected coordinate points, and the preliminary focus area coordinate points that cause the polygon to be concave are deleted. The convex hull generation algorithm traverses all the sorted preliminary focus area coordinate points, retains only the preliminary focus area coordinate points that form a convex polygon, and takes the finally connected closed polygon as the boundary polygon. The vertex coordinate set of the boundary polygon is combined with the preliminary focus area coordinate points to divide the preliminary focus area.

[0049] Based on the position of the preliminary focus area in the standardized coordinate system, it is analyzed by the minimum bounding rectangle method to obtain the focus area coordinates.

[0050] Specifically, the minimum bounding rectangle method traverses the horizontal preliminary focus area coordinate values of all preliminary focus area coordinate points in the preliminary focus area coordinate point set, records the minimum value of the left boundary horizontal coordinate value as the left boundary horizontal preliminary focus area coordinate value of the focus area, and records the maximum value of the right boundary horizontal preliminary focus area coordinate value as the right boundary horizontal preliminary focus area coordinate value of the focus area. The minimum bounding rectangle method traverses all the vertical boundary coordinate values of the coordinate points in the preliminary focused area coordinate set, records the minimum value of the vertical boundary coordinate value as the lower boundary vertical preliminary focused area coordinate value of the focused area, and records the maximum value of the vertical boundary coordinate value as the upper boundary vertical preliminary focused area coordinate value of the focused area. The left boundary horizontal preliminary focused area coordinate value, the right boundary horizontal preliminary focused area coordinate value, the lower boundary vertical coordinate value, and the upper boundary vertical coordinate value are combined into the four vertex coordinates of the focused area: the left boundary horizontal preliminary focused area coordinate value and the upper boundary vertical preliminary focused area coordinate value form the upper left vertex coordinate, the right boundary horizontal preliminary focused area coordinate value and the upper boundary vertical preliminary focused area coordinate value form the upper right vertex coordinate, the right boundary horizontal preliminary focused area coordinate value and the lower boundary vertical preliminary focused area coordinate value form the lower right vertex coordinate, and the left boundary horizontal preliminary focused area coordinate value and the lower boundary vertical preliminary focused area coordinate value form the lower left vertex coordinate. The horizontal coordinate values and vertical coordinate values of the four vertex coordinates are all represented by the values in the standardized coordinate system to obtain the focused area coordinates.

[0051] The instruction generation module fuses the initial light strategy of the scene and the focused area coordinates through a weighted fusion method, and combines the user's visual acuity parameters to form light parameters.

[0052] Based on the initial light strategy data of the scene and the focused area coordinates, the influence degree and correlation of the user's visual experience are analyzed through the principal component analysis method to obtain the influence weight of the user's visual experience, and the weighted fusion method is used for fusion to obtain the light strategy.

[0053] Specifically, the initial light strategy data of the scene and the focused area coordinate data are input into the principal component analysis method. The principal component analysis method performs standardized processing on the light intensity parameter, color temperature parameter of the initial light strategy data of the scene and the horizontal position parameter, vertical position parameter of the focused area coordinate data to eliminate the dimension difference. The principal component analysis method calculates the covariance values between two variables for the standardized light intensity parameter, color temperature parameter, horizontal position parameter, and vertical position parameter, and arranges all the pairwise covariance values of the four variables into a 4×4 symmetric matrix. The element in the i-th row and j-th column of the symmetric matrix is the covariance value between the i-th parameter and the j-th parameter, and the covariance matrix is obtained; The eigenvalue decomposition of the covariance matrix is used to extract the principal component direction and the variance contribution rate. The principal component direction corresponds to the eigenvector with the largest variance contribution rate, and the absolute values of the elements of the eigenvector with the largest variance contribution rate represent the influence weights of the light intensity parameter, color temperature parameter, horizontal position parameter, and vertical position parameter on the user's visual experience. The principal component analysis method normalizes the influence weights into weight coefficients with a sum of 1. The light intensity parameter and color temperature parameter of the initial light strategy data of the scene, and the horizontal position parameter and vertical position parameter of the focused area coordinate data perform a weighted summation operation according to the weight coefficients, and the weighted summation result is used as the fused light strategy parameter. The combination of the light intensity value and color temperature value in the light strategy parameter is used as the light strategy.

[0054] Combine the light strategy with the user's vision parameters, and through an adaptive adjustment algorithm, perform optimization and matching to form light parameters suitable for the user.

[0055] Specifically, the vision test report data generated by medical testing equipment, including the specific values of the vision level parameters measured by the optometer, the light sensitivity parameters recorded by the light sensitivity tester, and the color vision difference parameters calibrated by the color vision test card, are used to obtain the user's vision parameters; The light intensity parameter and color temperature parameter in the light strategy, and the vision level parameter, light sensitivity parameter, and color vision difference parameter in the user's vision parameters are input into the adaptive adjustment algorithm. The adaptive adjustment algorithm performs normalization processing on the user's vision parameters, converting the vision level parameter, light sensitivity parameter, and color vision difference parameter into values in the range of 0 to 1; the adaptive adjustment algorithm adjusts the light intensity parameter and color temperature parameter according to the normalized user's vision parameter values. The light intensity parameter and the vision level parameter values perform an inverse linear adjustment, and the color temperature parameter and the light sensitivity parameter values perform a forward linear adjustment. The adaptive adjustment algorithm inputs the adjusted light intensity parameter, color temperature parameter, and color vision difference parameter in the user's vision parameters into the color temperature compensation formula. The color temperature compensation formula performs an addition and subtraction offset operation on the color temperature parameter according to the color vision difference parameter value. The adaptive adjustment algorithm performs a range limit operation on the finally adjusted light intensity parameter and color temperature parameter. The light intensity parameter is limited between the minimum and maximum values allowed by the user's vision parameters, for example, the minimum value of 3000K warm yellow light and the maximum value of 6000K cold white light. The color temperature parameter is limited within the color temperature range defined by the user's comfort level. The combined light intensity parameter and color temperature parameter after the range limit are used as the light parameters suitable for the user.

[0056] Analyze through the NSGA-II algorithm to obtain the lighting control instruction.

[0057] Based on the light parameters, set the multi-objective functions of lighting energy saving and visual comfort, and randomly generate a population of lighting control instruction parameter combinations containing multiple individuals and initialize it.

[0058] Specifically, taking the allowable range of the illumination intensity parameter and the allowable range of the color temperature parameter in the optical parameters as constraint conditions, the lighting energy-saving objective function is set as the ratio of the lamp energy consumption value to the rated power, and the visual comfort objective function is the sum of the absolute deviation between the actual illumination intensity parameter and the illumination intensity parameter in the optical parameters and the absolute deviation between the actual color temperature parameter and the color temperature parameter in the optical parameters. The random number generator generates a random illumination intensity parameter value within the allowable range of the illumination intensity parameter, generates a random color temperature parameter value within the allowable range of the color temperature parameter, and generates a random lamp switching time parameter value within the allowable range of the lamp switching time. The three random parameter values are combined into an individual of the lighting control instruction parameter combination. The random parameter value generation operation is repeated to generate a preset number of individuals of the lighting control instruction parameter combination, and all individuals constitute the initialized population of the lighting control instruction parameter combination.

[0059] Perform non-dominated sorting and crowding degree analysis on the population of the lighting control instruction parameter combination, through selection, crossover, mutation, merging, updating, and elitist retention.

[0060] Specifically, each individual in the population of the lighting control instruction parameter combination is input into the non-dominated sorting method. The non-dominated sorting method compares the values of the individual in the lighting energy-saving objective function and the visual comfort objective function to judge the dominance relationship between individuals; the non-dominated sorting method divides the individuals without a dominated relationship into the first non-dominated rank, and the remaining individuals are compared for the dominance relationship again and divided into the second non-dominated rank, and the operation is repeated until all individuals are assigned non-dominated ranks; the crowding degree analysis method performs an evaluation of the spatial distribution density of the objective function for the individuals within the same non-dominated rank, and measures the sum of the differences in the lighting energy-saving objective function value and the visual comfort objective function value between each individual and its adjacent individuals as the crowding degree value. The selection operation filters and retains individuals with high ranks and high crowding degrees according to the non-dominated rank priority and the crowding degree value. The crossover operation performs arithmetic averaging on the illumination intensity parameter value, the color temperature parameter value, and the lamp switching time parameter value of two individuals to generate new individuals. The mutation operation applies random perturbations to the illumination intensity parameter value, the color temperature parameter value, and the lamp switching time parameter value of the individual to generate new individuals; the merging operation combines the retained original individuals with the newly generated crossover individuals and mutation individuals into a temporary population. The elitist retention operation extracts the top 10% of the individuals in the non-dominated rank from the original population and directly replaces the same number of the lowest-rank individuals in the temporary population. The updating operation re-performs non-dominated sorting and crowding degree analysis on the temporary population, and screens out individuals with the same number as the original population as the new generation of the lighting control instruction parameter combination population.

[0061] When the multi-objective functions of lighting energy saving and visual comfort converge, select the optimal individual from the population of the lighting control instruction parameter combination to obtain the lighting control instruction.

[0062] Specifically, by analyzing the absolute value of the change rate of the lighting energy-saving objective function value and the absolute value of the change rate of the visual comfort objective function value recorded during the iteration of the historical lighting control instruction parameter combination population, the 95% quantiles of the two sets of values are respectively taken as the initial candidate values. After verification and adjustment by the test data set, the final double-objective convergence criterion threshold range is determined. When the change rates of the lighting energy-saving objective function and the visual comfort objective function of the final lighting control instruction parameter combination population are both less than the double-objective convergence criterion threshold, it is determined to converge; Traverse each individual in the final lighting control instruction parameter combination population, check whether the non-dominated rank of the individual is the first non-dominated rank, and retain all individuals with the first non-dominated rank to form a candidate optimal individual set. The candidate optimal individual set is input into the crowding degree analysis method, and the crowding degree value of each candidate optimal individual in the lighting energy-saving objective function and visual comfort objective function space is calculated, and the candidate optimal individuals are sorted from largest to smallest according to the crowding degree value; the candidate optimal individual with the largest crowding degree value is selected as the optimal individual, and the coincidence degree of the light intensity parameter value of the optimal individual with the allowable range of the light intensity parameter in the light parameters and the coincidence degree of the color temperature parameter value with the allowable range of the color temperature parameter in the light parameters are verified. When the coincidence degree exceeds 90%, the optimal individual is confirmed to be valid. The combination of the light intensity parameter value, color temperature parameter value, and lamp switch time parameter value of the optimal individual is used as the lighting control instruction.

[0063] The execution module transmits and executes based on the lighting control instruction through a wireless communication method, and real-time monitors the user's visual physiological response to obtain feedback data.

[0064] Encode the lighting control instruction, convert it into the JSON format for wireless communication transmission, and send it to the lighting end through the wireless communication protocol.

[0065] Specifically, the light intensity parameter value, color temperature parameter value, and lamp switch time parameter value in the lighting control instruction are arranged in the key-value pair format. The light intensity parameter value retains two decimal places, the color temperature parameter value takes an integer, and the lamp switch time parameter value is converted into a string in the "HH:MM" format. The key-value pairs are encapsulated into a JSON format data packet. The wireless communication protocol defines that the data packet structure includes a target device address field, an instruction type field, a JSON format data field, and a check code field. The check code field is generated by executing the CRC32 algorithm. The wireless communication protocol converts the complete data packet into a binary byte stream, establishes a Socket connection with the lighting end through the TCP / IP protocol, and the byte stream is fragmented and encapsulated into TCP data frames. Each TCP data frame is added with a frame header identifier and a frame tail check sequence, and is continuously sent to the lighting end at an interval of 500 ms. After the sending is completed, wait for the lighting end to return an acknowledgment frame. When the acknowledgment frame is not received, start a 3-second timeout retransmission mechanism, and the maximum number of retransmissions is set to 3 times.

[0066] The lighting terminal receives the lighting control instruction transmitted by wireless communication, decodes and restores it to the original lighting control instruction, and adjusts and executes the lighting parameters according to the lighting control instruction.

[0067] Specifically, the lighting terminal receives the binary byte stream data packet transmitted by wireless communication through the TCP / IP protocol. The binary byte stream data packet performs integrity verification according to the frame header identifier and the frame tail check sequence, and the check sequence is generated by the CRC32 algorithm. After the integrity verification passes, the binary byte stream data packet parses the target device address field and the instruction type field, extracts the JSON format data field and converts it into minutes after matching the current lighting terminal address and the instruction type. The value of the light intensity parameter is compared with the allowable range of light intensity in the light parameters, and the value of the color temperature parameter is compared with the allowable range of color temperature in the light parameters. When it exceeds the allowable range, the boundary value of the allowable range is used for substitution. The lighting terminal generates a pulse width modulation signal to adjust the lamp drive current, converts the value of the light intensity parameter into a duty cycle value and outputs a color temperature control voltage signal, converts the value of the color temperature parameter into a voltage amplitude and outputs it. The lamp switch time parameter value is input into the timer unit, and the timer unit is configured in the countdown mode. After the countdown ends, the relay switch action is triggered. After the lighting terminal executes the parameter adjustment, it returns a confirmation frame containing the execution result code to the sending end through the TCP / IP protocol. The execution result code includes the actual value of the light intensity, the actual value of the color temperature, and the remaining seconds of the switch time.

[0068] The user's visual physiological response data is monitored in real time and preprocessed to extract the feedback data related to the user's visual physiological response.

[0069] Specifically, the user's visual physiological response is monitored in real time to extract the pupil and iris boundaries, and the physiological index of the user's pupil diameter change rate is calculated. The near-infrared spectrum sensor is used to record the user's blink frequency data at a sampling frequency of 30 times per second, and the eye movement tracking device is used to obtain the coordinate data of the area where the user's line of sight is fixed for a long time at a refresh rate of 120Hz. The original physiological index of the user's pupil diameter change rate is processed by moving average filtering, and the window width is set to 10 sampling points to remove high-frequency noise. The original user blink frequency data is processed by identifying the characteristics of the closed-eye action. According to the characteristic that the eyelid coverage exceeds 80% and the duration is greater than 100ms, the closed-eye action is identified. The original coordinate data of the area where the user's line of sight is fixed for a long time is processed by removing outliers, and the coordinate points deviating from the mean by three standard deviations are removed. The average value and variance per minute are extracted from the preprocessed physiological index of the user's pupil diameter change rate as feedback data. The number of blinks per minute and the average duration are extracted from the preprocessed user blink frequency data as feedback data. The cluster center coordinates and distribution radius are extracted from the preprocessed coordinate data of the area where the user's line of sight is fixed for a long time as feedback data.

[0070] Optimization module, which dynamically corrects optical parameters using a fuzzy logic algorithm based on the user's fatigue state to obtain optimized optical parameters.

[0071] The user's fatigue state is fuzzified through a fuzzy logic algorithm, transformed into a fuzzy linguistic variable, and logical reasoning is performed through a fuzzy inference method in combination with a fuzzy rule base to obtain a fuzzy conclusion for adjusting optical parameters.

[0072] Specifically, the user's fatigue state is input into the fuzzy logic algorithm. The fuzzy logic algorithm defines the fuzzy linguistic variables of the user's fatigue state as "fatigue" and "non-fatigue", corresponding to triangular membership functions. Exemplarily, the vertices of the triangular membership functions are respectively set to 0.8 times, 1.0 times, and 1.2 times the fatigue threshold. By collecting the user's visual physiological response data under different combinations of light intensity and color temperature, with the amount of user's visual physiological response data covering 50 groups, the Pearson coefficient is used to screen the user's visual physiological response characteristics with statistical relevance for the percentage of optical parameter adjustment amount, and an initial rule base is formed by combining logical relationships; the rule consequent parameters are adjusted through the training set (70%), and the classification accuracy is verified through the test set (30%). After iterative optimization until the accuracy exceeds 85%, finally 8 - 12 fuzzy rules are generated, such as the exemplary "High pupil index + Low blink variance + Medium distribution radius → Light intensity + 25%, Color temperature - 15%". The fuzzy inference method uses the Mamdani inference model, performs minimum-maximum operations on the antecedent membership degree and consequent membership degree of each fuzzy rule. After aggregating the output membership functions of all fuzzy rules, defuzzification is performed through the centroid method to generate a fuzzy conclusion for adjusting optical parameters. Exemplary fuzzy conclusions for adjusting optical parameters include the adjustment direction (increase / decrease / maintain) and percentage of adjustment amount (10% - 30%) of light intensity, and the adjustment direction (increase / decrease / maintain) and percentage of adjustment amount (5% - 20%) of color temperature.

[0073] The centroid method is used to defuzzify the fuzzy conclusion, which is transformed into a numerical value to obtain the optimized optical parameters.

[0074] Specifically, input the light intensity adjustment direction (increase / decrease / maintain) and adjustment percentage (10%-30%) of the fuzzy conclusion of light parameter adjustment, and the color temperature adjustment direction (increase / decrease / maintain) and adjustment percentage (5%-20%) into the centroid method. The membership function of the light intensity adjustment percentage is discretely sampled at a step of 0.1% in the exemplary domain [-30%, 30%], and the membership function of the color temperature adjustment percentage is discretely sampled at a step of 0.1% in the exemplary domain [-20%, 20%]. Multiply the abscissa value of each discrete point by the corresponding ordinate membership value, sum up all the product results of the light intensity adjustment percentage, and divide by the sum of the membership degrees. Sum up all the product results of the color temperature adjustment percentage and divide by the sum of the membership degrees. Round the weighted average of the light intensity adjustment percentage to one decimal place, and round the weighted average of the color temperature adjustment percentage to one decimal place. Multiply the light intensity value of the current light parameter by 1 + light intensity adjustment percentage / 100 to obtain the optimized light intensity value, multiply the color temperature value of the current light parameter by 1 + color temperature adjustment percentage / 100 to obtain the optimized color temperature value, and combine the optimized light intensity value and the optimized color temperature value into the optimized light parameter.

[0075] Combined with the feedback data, generate the optimal intelligent eye protection lighting control parameters through data fusion and intelligent decision-making algorithms.

[0076] Combine the optimized light parameters with the feedback data, perform integration processing through weighted average fusion, and conduct in-depth analysis by using Bayesian decision-making to generate the optimal intelligent eye protection lighting control parameters.

[0077] Specifically, input the light intensity value and color temperature value in the optimized light parameters, the mean physiological index of the user's pupil diameter change rate and the data variance of the user's blink frequency in the feedback data into the weighted average fusion method. The weighted average fusion method sets the weight coefficient of the light intensity value according to the historical fluctuation range of the mean physiological index of the user's pupil diameter change rate in the feedback data, and sets the weight coefficient of the color temperature value according to the historical fluctuation range of the data variance of the user's blink frequency. The Bayesian decision-making method establishes the prior probability distribution of the light intensity value and the color temperature value, fuses the light intensity value and the fused color temperature value, inputs them into the Bayesian decision-making method to analyze the posterior probability distribution. The Bayesian decision-making method traverses all possible combinations of the light intensity value and the color temperature value, selects the light intensity value and the color temperature value with the largest posterior probability as the intelligent eye protection lighting control parameters, and verifies whether the light intensity value is within the range allowed by the user's visual parameters and whether the color temperature value is within the color temperature range defined by the user's comfort level. The finally verified combination of the light intensity value and the color temperature value is the intelligent eye protection lighting control parameters.

[0078] In summary, the present invention can: predict environmental changes through the Markov model, be able to perceive in advance the evolution trend of factors such as environmental light intensity, adjust the light strategy in advance, and avoid the lighting lagging behind environmental changes; improve the lighting forward-looking and adaptability, reduce the impact of environmental mutations on the lighting effect, create a more stable and comfortable lighting environment for users, and enhance the user experience; the fuzzy logic algorithm responds to the fatigue state, and can convert the complex and fuzzy fatigue state of users into precise light parameter adjustment instructions; it can flexibly adapt to the lighting environment according to different fatigue degrees, effectively relieve visual fatigue, provide a personalized and dynamic eye protection lighting solution for users, and improve the lighting intelligence and user comfort.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An intelligent eye protection lighting control system, characterized in that: including a data acquisition module, which collects multi-source data in real time for preprocessing, analyzes it through a Markov model, and obtains an initial light strategy for the scene; a physiological analysis module, which calculates the physiological index of the user's pupil diameter and marks fatigue warnings based on the preprocessed multi-source data through the Daugman algorithm to obtain the user's fatigue state. According to the area where the user's line of sight is fixed for a long time, through spatial coordinate mapping and clustering algorithms, the line-of-sight landing point is processed and the area is divided to generate the coordinates of the focused area; an instruction generation module, which fuses the initial light strategy of the scene and the coordinates of the focused area through a weighted fusion method, combines the user's vision parameters to form light parameters, and analyzes them through the NSGA-II algorithm to obtain lighting control instructions; an execution module, which transmits and executes based on the lighting control instructions through a wireless communication method, and monitors the user's visual physiological reactions in real time to obtain feedback data; an optimization module, which dynamically corrects the light parameters based on the user's fatigue state using a fuzzy logic algorithm to obtain optimized light parameters, and combines the feedback data through a data fusion and intelligent decision-making algorithm to generate optimal intelligent eye protection lighting control parameters.

2. The intelligent eye protection lighting control system according to claim 1, wherein: The multi-source data includes environmental light intensity data, environmental color temperature data, user pupil diameter data, user blink frequency data, and data on the area where the user's line of sight is fixed for a long time; The preprocessing includes standardization, noise removal, filling missing values, and outlier processing.

3. The intelligent eye protection lighting control system according to claim 2, characterized in that: The analysis through the Markov model to obtain the initial light strategy of the scene is as follows Based on the preprocessed historical multi-source data, the Markov model is trained through the maximum likelihood estimation method to obtain the trained Markov model; The preprocessed multi-source data is input into the Markov model. According to the transition probability relationship between different environmental and user state data, the change trend of environmental light intensity in the current scene is predicted, and the corresponding preliminary light strategy parameters in different scenes are analyzed; The preliminary light strategy parameters are sorted and screened through the analytic hierarchy process to obtain the initial light strategy of the scene.

4. The intelligent eye protection lighting control system according to claim 1, characterized in that: The calculation of the user's physiological index and marking of fatigue warnings based on the preprocessed multi-source data through the Daugman algorithm to obtain the user's fatigue state is as follows The preprocessed multi-source data is input into the Daugman algorithm. Through circle detection and bilinear interpolation, the pupil and iris boundaries are extracted, and the physiological index of the change rate of the user's pupil diameter is calculated; Based on historical fatigue data and historical normal state data, the parameters are optimized through the support vector machine method to obtain the fatigue threshold; The physiological index of the change rate of the user's pupil diameter and the user's blink frequency data are compared with the fatigue threshold to obtain the user's fatigue state.

5. The intelligent eye protection lighting control system according to claim 1, wherein: The generation of the coordinates of the focused area by processing the line-of-sight landing point and dividing the area according to the area where the user's line of sight is fixed for a long time through spatial coordinate mapping and clustering algorithms is as follows The area where the user's line of sight is fixed for a long time is converted into a unified standardized coordinate system through the spatial coordinate mapping algorithm; Through the DBSCAN clustering algorithm, clustering is performed according to the spatial position similarity of the area where the user's line of sight is fixed for a long time, and the preliminary focused area is divided, and abnormal landing points are removed and the boundary is smoothed; Based on the position of the preliminary focus area in the standardized coordinate system, analyze it by the minimum bounding rectangle method to obtain the coordinates of the focus area.

6. The intelligent eye protection lighting control system according to claim 5, wherein: Fuse the initial scene light strategy and the coordinates of the focus area through a weighted fusion method, and combine the user's visual acuity parameters to form light parameters. The specific steps are as follows: Based on the initial scene light strategy data and the coordinates of the focus area, analyze the influence degree and correlation on the user's visual experience through the principal component analysis method, obtain the influence weight of the user's visual experience, and use the weighted fusion method for fusion to obtain the light strategy. Combine the light strategy with the user's visual acuity parameters, and through an adaptive adjustment algorithm, optimize and match them to form light parameters.

7. The intelligent eye protection lighting control system according to claim 6, wherein: Analyze through the NSGA-II algorithm to obtain the lighting control instruction. The specific steps are as follows: Based on the light parameters, set the multi-objective functions of lighting energy saving and visual comfort, randomly generate a population of lighting control instruction parameter combinations containing multiple individuals and initialize them. Perform non-dominated sorting and crowding degree analysis on the population of lighting control instruction parameter combinations, and through selection, crossover, mutation, combination, update, and elitist retention. When the multi-objective functions of lighting energy saving and visual comfort converge, select the optimal individual from the population of lighting control instruction parameter combinations to obtain the lighting control instruction.

8. The intelligent eye protection lighting control system according to claim 7, wherein: Transmit and execute based on the lighting control instruction through a wireless communication method, and monitor the user's visual physiological reaction in real time to obtain feedback data. The specific steps are as follows: Encode the lighting control instruction, convert it into the JSON format for wireless communication transmission, and send it to the lighting end through the wireless communication protocol. The lighting end receives the lighting control instruction transmitted wirelessly, decodes and restores it to the original lighting control instruction, and adjusts and executes the lighting parameters according to the lighting control instruction. Monitor the user's visual physiological reaction data in real time and perform preprocessing to extract the feedback data related to the user's visual physiological reaction.

9. The intelligent eye protection lighting control system according to claim 1, wherein: Dynamically correct the light parameters based on the user's fatigue state using the fuzzy logic algorithm to obtain the optimized light parameters. The specific steps are as follows: Fuzzify the user's fatigue state through the fuzzy logic algorithm, transform it into a fuzzy linguistic variable, and perform logical reasoning through the fuzzy inference method in combination with the fuzzy rule base to obtain the fuzzy conclusion of the light parameter adjustment. Use the centroid method to defuzzify the fuzzy conclusion and transform it into a numerical value to obtain the optimized light parameters.

10. The intelligent eye protection lighting control system according to claim 1, wherein: Combine the optimized light parameters with the feedback data, perform integration processing through weighted average fusion, and conduct in-depth analysis using Bayesian decision-making to generate the optimal intelligent eye protection lighting control parameters.

Images