Children myopia prevention and control device based on light environment regulation

Abstract

The application discloses a myopia prevention and control device based on light environment regulation, and belongs to the technical field of myopia prevention. The application comprises a device main body, a central processing control unit, a multi-dimensional light environment sensing unit, a dynamic light environment adjusting unit, a child eye use behavior monitoring unit and a hierarchical early warning interaction unit. Key light parameters such as light intensity and color temperature are collected in real time, data such as eye distance and eye use posture are synchronously acquired, adaptive adjustment instructions are generated according to child myopia prevention standards, and the dynamic light environment adjusting unit is driven to optimize light. Starting from the two dimensions of environmental adaptation and behavior correction, the eye burden is accurately reduced, the occurrence and development of myopia are effectively prevented, target eye use data containing data tags are acquired, the visual line deviation habits and standard reference habits of children are determined, abnormal postures are identified, the children's bad eye use postures are corrected, eye use scenes are identified, optimal light environment parameters are extracted according to the scenes, and the pertinence and efficiency of light environment regulation are improved.

Description

Technical Field

[0001] This invention relates to the field of myopia prevention technology, and in particular to a device for myopia prevention and control in children based on light environment regulation. Background Technology

[0002] Currently, children commonly face problems such as insufficient light intensity, unsuitable color temperature, significant glare, and uneven illuminance distribution in their learning and daily lives. Prolonged exposure to such environments exacerbates eye fatigue and accelerates the onset and progression of myopia. Most eye-protection lighting products only provide fixed-parameter illumination and cannot dynamically adjust according to children's eye-use scenarios, failing to meet the lighting environment needs of different types of work. Furthermore, existing myopia prevention methods often focus on single-dimensional interventions, lacking coordinated monitoring and comprehensive control of light environment parameters and eye-use behavior, and cannot provide real-time feedback to parents on children's eye health, resulting in insufficiently targeted and limited-effect interventions. Summary of the Invention

[0003] The purpose of this invention is to provide a device for the prevention and control of myopia in children based on light environment regulation, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a children's myopia prevention and control device based on light environment regulation, comprising a device body, a central processing and control unit, a multi-dimensional light environment sensing unit, a dynamic light environment adjustment unit, a children's eye use behavior monitoring unit, a graded early warning and interaction unit, a data storage unit, and a power supply unit;

[0005] The multi-dimensional light environment sensing unit is used to collect light environment parameters in children's learning and living spaces in real time;

[0006] The child eye behavior monitoring unit is used to acquire the child's target eye data in a non-contact manner, determine whether the line of sight deviates from the normal eye angle, and generate eye posture warning information.

[0007] The central processing and control unit is used to perform fusion analysis and logical judgment on the collected light environment parameters and eye behavior data, generate adaptive adjustment instructions according to the optical standards for myopia prevention in children, drive the dynamic light environment adjustment unit to output appropriate lighting light, and optimize the local eye light environment in real time.

[0008] The graded early warning interaction unit is used to execute graded reminders when the light environment is not up to standard or when eye use behavior is abnormal.

[0009] The data storage unit is used to record control data and eye habits; the power supply unit is used to provide working power.

[0010] Furthermore, the multi-dimensional light environment sensing unit includes a high-precision light intensity sensor, a color temperature acquisition sensor, a blue light sensor, a color rendering index detection module, a glare detection module, and an illuminance uniformity analysis module. The multi-dimensional light environment sensing unit collects light environment parameters and transmits them to the central processing and control unit through a time-division synchronous acquisition mechanism.

[0011] Furthermore, the child's eye behavior monitoring unit adopts a combination of an infrared ranging module and a high-definition image recognition module. It uses a binocular visual positioning algorithm to monitor the real-time object-eye distance between the child's eyes and the visual target, while recording the duration of continuous eye use, identifying the real-time eye posture and determining whether it is an abnormal posture, and judging whether the line of sight deviates from the normal eye use angle.

[0012] Furthermore, the child's eye behavior monitoring unit determines whether the line of sight deviates from the normal eye angle, specifically including:

[0013] Based on the child eye use behavior monitoring unit, the target eye use data for monitoring the child's eyes is determined. The target eye use data includes real-time object-eye distance, continuous eye use duration and real-time eye use posture. At the same time, the data tags of the target eye use data are obtained.

[0014] The data tags of the target eye use data are input into a preset eye use standard library for matching, and the benchmark eye use type for standardizing the judgment of children's eye posture is output based on the matching results. The preset eye use standard library includes benchmark eye use types for looking straight ahead, reading, writing, looking into the distance, and watching screens.

[0015] The target eye use data is matched item by item with the benchmark eye use type, and the target eye use data is initially classified based on the matching results to generate a first type of eye use data segment and a second type of eye use data segment. The first type of eye use data segment consists of the target eye use data that does not match the benchmark eye use type, and the second type of eye use data segment consists of the target eye use data that matches the benchmark eye use type.

[0016] Based on the matching results, the data in the second category of eye use data segment are further classified according to the baseline eye use type to generate multiple sets of standardized eye use data groups.

[0017] Randomly select several data points from the first type of eye use data segment as the first analysis sample, and at the same time, randomly select data from each group of standardized eye use data groups as the second analysis sample;

[0018] The first analysis sample is analyzed to determine the gaze feature points of the child's eyes and head. At the same time, a real-time gaze trajectory network of the child is constructed based on the gaze feature points, and the first type of eye use data is deployed to the real-time gaze trajectory network based on the gaze feature points.

[0019] Based on the deployment results, the distribution characteristics of the first type of eye use data in the real-time gaze trajectory network are determined, and the children's real-time gaze deviation habits are determined based on the distribution characteristics.

[0020] A standard gaze reference network is constructed based on the aforementioned baseline eye use type. Simultaneously, the second analysis sample is sequentially input into the standard gaze reference network for posture fitting, and the child's standard gaze reference habits are determined based on the fitting results.

[0021] Obtain the first weight of real-time gaze offset habits and the second weight of standardized gaze reference habits. At the same time, score the real-time gaze offset habits according to the first weight to obtain the first score, and score the standardized gaze reference habits according to the second weight to obtain the second score.

[0022] A comprehensive score for children's eye use is calculated by weighting the first and second scores, and the comprehensive score is compared with a preset scoring threshold to determine whether the child's line of sight deviates from the normal eye use angle.

[0023] When the overall score is greater than or equal to the preset score threshold, the child's line of sight is determined to be normal and their eye posture is correct.

[0024] Otherwise, the system determines that the child's gaze angle is abnormal and deviates from the standard eye use direction, and generates an eye use posture warning when the gaze deviation determination is valid.

[0025] Furthermore, the central processing control unit has a built-in threshold library for the light environment for myopia prevention in children and a standard library for safe eye use behavior, which includes safe eye use distance thresholds, maximum continuous eye use duration thresholds, suitable light intensity ranges, standard color temperature ranges, and glare limit values ​​for children of different age groups.

[0026] The central processing and control unit has four preset eye-use scenario modes: learning and reading, using electronic devices, reading picture books, and resting and relaxing. It automatically matches the optimal combination of light environment parameters and generates precise light environment adjustment instructions, specifically including:

[0027] It receives and analyzes the target eye use data uploaded by the eye use behavior monitoring unit and the judgment results of whether the line of sight deviates from the normal eye use angle in real time;

[0028] By comparing the target eye use data with the eye use behavior safety standard database, the type of eye use scenario the child is currently in can be determined and the corresponding scenario can be identified.

[0029] Extract the preset target illumination intensity parameters, target color temperature parameters, target color rendering index parameters, and glare suppression parameters for this scene;

[0030] The difference between the current real-time light environment parameters and the target light environment parameters is calculated to obtain the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient.

[0031] Based on the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient, a light environment adjustment command containing the adjustment range, adjustment speed, and adjustment duration is generated and sent to the dynamic light environment adjustment unit for execution.

[0032] Furthermore, the type of eye-use scenario the child is currently in is determined and the corresponding scenario is identified, specifically including:

[0033] Acquire target eye use data uploaded by the children's eye use behavior monitoring unit, and extract current light environment parameters synchronously collected by the multi-dimensional light environment perception unit;

[0034] The built-in eye behavior safety standard library is invoked to extract the scene feature parameter set corresponding to the preset scene. The scene feature parameter set includes the standard object-eye distance range, typical eye posture range, characteristic continuous eye use duration threshold and adaptive light environment parameter range for each scene.

[0035] Standardize and preprocess the target eye use data and current light environment parameters, remove abnormal fluctuation data points, and generate a normalized effective data set.

[0036] Construct a scene matching feature vector by mapping the real-time object-eye distance, continuous eye use duration, real-time eye use posture, gaze angle state, ambient light intensity and color temperature parameters in the effective dataset to vector dimensions, and generate the feature vector to be matched.

[0037] Calculate the cosine similarity between the feature vector to be matched and the standard feature vectors corresponding to four preset scene feature parameter sets to obtain four similarity coefficients, the value of which ranges from 0 to 1;

[0038] Set a similarity threshold of 0.75, filter out preset scenes with similarity coefficients greater than or equal to the threshold as candidate scenes, and then filter and lock the final eye-use scene type from the candidate scenes.

[0039] Furthermore, the final eye-use scenario type is selected and locked from the candidate scenarios, specifically including:

[0040] After filtering out candidate scenarios, if there is only one candidate scenario, the candidate scenario is determined to be the type of eye-use scenario in which the child is currently experiencing and is locked.

[0041] If there are at least two candidate scenarios, extract the differential feature parameters of the candidate scenarios. The differential feature parameters include screen blue light radiation correlation parameters specific to electronic device usage scenarios, close-range viewing duration parameters specific to picture book viewing scenarios, writing action correlation parameters specific to learning and reading scenarios, and the proportion of time spent looking into the distance specific to rest and relaxation scenarios.

[0042] A secondary matching scoring model is constructed based on differential feature parameters to score each candidate scenario, with each item scored from 0 to 25 points, for a total score of 100 points.

[0043] Substitute the corresponding parameters from the valid dataset into the secondary matching scoring model, calculate the final score for each candidate scenario, and select the candidate scenario with the highest score as the current eye-use scenario type for the child and lock it in.

[0044] If the similarity coefficient of all preset scenes is less than 0.75, the default scene matching mechanism will be activated. Based on the standard eye use characteristics corresponding to the child's current age, the closest preset scene will be matched as the temporary locked scene, and the target eye use data and light environment parameters will be continuously collected in the next 30 seconds for dynamic verification.

[0045] If the similarity coefficient of the temporarily locked scene during the dynamic verification period is consistently greater than or equal to 0.75, the temporarily locked scene will be determined as the final eye-use scene type and locked; if the similarity coefficient is still less than 0.75, the scene abnormality prompt information will be output to the hierarchical warning interaction unit, and the temporarily locked scene will be maintained until the data meets the matching conditions.

[0046] Furthermore, the dynamic light environment adjustment unit includes a full-spectrum LED light source array, a PWM pulse width adjustment module, a color temperature continuous adjustment module, a glare suppression module, and a light diffusion module, and the dynamic light environment adjustment unit is disposed on the main body of the device.

[0047] Furthermore, the main body of the device is equipped with a multi-functional installation and fixing unit, which includes three installation structures: a clamping base, a wall-mounted bracket, and a desktop support.

[0048] Furthermore, the graded early warning interaction unit includes a voice prompt module, an LED status indicator module, a vibration reminder module, and a text display module. The graded early warning is divided into three levels of response based on the risk level: mild reminder, moderate warning, and severe alarm. At the same time, abnormal information, eye posture warning information, and control records are simultaneously pushed to the parent terminal device via wireless communication.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. This invention achieves integrated analysis of light environment and eye behavior through the collaboration of a multi-dimensional light environment sensing unit, a child eye behavior monitoring unit, and a central processing control unit. It can collect key light parameters such as light intensity and color temperature in real time, and simultaneously acquire data such as object-eye distance and eye posture. Based on the standards for myopia prevention in children, it generates adaptive adjustment instructions to drive the dynamic light environment adjustment unit to optimize the light. By addressing both environmental adaptation and behavioral correction, it precisely reduces the burden on children's eyes and effectively prevents the occurrence and development of myopia. The multi-functional installation and fixing unit is adaptable to various locations such as desks and bedside tables, ensuring that light covers key eye areas and comprehensively improving the safety and convenience of children's eye environment.

[0051] 2. This invention acquires target eye use data with data tags, matches it with a preset eye use standard library to determine the baseline eye use type, and identifies children's gaze deviation habits and standard reference habits. It can accurately identify abnormal postures such as forward head tilt and side tilt, helping to correct children's poor eye use posture in a timely manner and reduce the risk of myopia. The central processing control unit has a built-in scene matching model that can accurately identify eye use scenarios and achieve precise scene locking. Based on the scenario, it extracts the optimal light environment parameters and generates precise adjustment instructions, so that the light parameters are adapted to the needs of different scenarios. For example, the electronic device scenario is adapted to blue light filtering parameters, and the rest scenario triggers a distance viewing reminder, improving the pertinence and efficiency of light environment control. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the process for preventing and controlling myopia in children based on light environment regulation according to the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1 The present invention provides the following technical solutions:

[0055] The device for preventing and controlling myopia in children based on light environment regulation includes the main body of the device, a central processing and control unit, a multi-dimensional light environment sensing unit, a dynamic light environment adjustment unit, a children's eye use behavior monitoring unit, a graded early warning and interaction unit, a data storage unit, and a power supply unit.

[0056] The multi-dimensional light environment sensing unit is used to collect light environment parameters in the children's learning and living space in real time. The light environment parameters include light intensity, light color temperature, light color rendering index, light uniformity, glare index, and environmental illuminance distribution parameters. The multi-dimensional light environment sensing unit includes a high-precision light intensity sensor, a color temperature acquisition sensor, a blue light sensor, a color rendering index detection module, a glare detection module, and an illuminance uniformity analysis module. The multi-dimensional light environment sensing unit collects light environment parameters and transmits them to the central processing and control unit through a time-division synchronous acquisition mechanism.

[0057] The child eye behavior monitoring unit is used to acquire the child's target eye data in a non-contact manner, determine whether the line of sight deviates from the normal eye angle, and generate eye posture warning information.

[0058] The central processing and control unit is used to perform fusion analysis and logical judgment on the collected light environment parameters and eye behavior data, generate adaptive adjustment instructions according to the optical standards for myopia prevention in children, drive the dynamic light environment adjustment unit to output appropriate lighting light, and optimize the local eye light environment in real time.

[0059] The graded warning interaction unit is used to execute graded reminders when the light environment is not up to standard or when eye use behavior is abnormal. The graded warning interaction unit includes a voice prompt module, an LED status indicator module, a vibration reminder module and a text display module. The graded warning is divided into three levels of response according to the risk level: mild reminder, moderate warning and severe alarm. At the same time, abnormal information, eye use posture warning information and control records are pushed to the parent terminal device through wireless communication.

[0060] The data storage unit is used to record control data and eye habits; the power supply unit is used to provide working power.

[0061] The dynamic light environment adjustment unit includes a full-spectrum LED light source array, a PWM pulse width adjustment module, a color temperature continuous adjustment module, a glare suppression module, and a light diffusion module. The light source covers the 380nm-780nm visible light band and can be steplessly adjusted within the color temperature range of 3000K to 6500K. The color rendering index Ra≥95. The dynamic light environment adjustment unit is set on the main body of the device.

[0062] The main body of the device is equipped with a multi-functional installation and fixing unit, which includes three installation structures: a clamp-type base, a wall-mounted bracket, and a desktop support base. It can be adapted to various installation positions such as desks, bedside tables, and bookshelves. After installation, it automatically calibrates the light illumination angle and monitoring angle to ensure that the light environment covers the main areas where children use their eyes.

[0063] In the above embodiments, the children's myopia prevention and control device integrates multiple module functions. The multi-dimensional light environment sensing unit can comprehensively collect key light parameters and ensure data accuracy through time-sharing synchronous collection. The children's eye behavior monitoring unit adopts non-contact monitoring, combining infrared ranging and image recognition to accurately acquire data on object-eye distance, eye usage time, and posture. The central processing and control unit integrates and analyzes the two types of data and generates adaptive adjustment instructions according to different scenarios. The full-spectrum LED light source of the dynamic light environment adjustment unit can achieve stepless color temperature adjustment, has a high color rendering index, and can provide high-quality light. The graded early warning and interaction unit provides graded reminders through multiple methods and can also synchronize information to parents, allowing parents to keep track of the situation in real time. The multi-functional installation and fixing unit is adaptable to various installation positions to ensure light coverage of key areas. The data storage unit records relevant data to help with subsequent analysis of children's eye habits. The overall device effectively prevents and controls children's myopia by addressing both light environment regulation and eye behavior standardization.

[0064] The child eye behavior monitoring unit uses a combination of an infrared ranging module and a high-definition image recognition module. It uses a binocular visual positioning algorithm to monitor the real-time object-eye distance between the child's eyes and the visual target, which includes books and electronic screens. It also records the duration of continuous eye use, identifies real-time eye posture, and determines whether it is an abnormal posture. Abnormal postures include forward head tilt, side tilt, and excessive head drooping.

[0065] Based on the child eye use behavior monitoring unit, the target eye use data for monitoring the child's eyes is determined. The target eye use data includes real-time object-eye distance, continuous eye use duration and real-time eye use posture. At the same time, the data tags of the target eye use data are obtained.

[0066] The data tags of the target eye use data are input into a preset eye use standard library for matching, and the benchmark eye use type for standardizing the judgment of children's eye posture is output based on the matching results. The preset eye use standard library includes benchmark eye use types for looking straight ahead, reading, writing, looking into the distance, and watching screens.

[0067] The target eye use data is matched item by item with the benchmark eye use type, and the target eye use data is initially classified based on the matching results to generate a first type of eye use data segment and a second type of eye use data segment. The first type of eye use data segment consists of the target eye use data that does not match the benchmark eye use type, and the second type of eye use data segment consists of the target eye use data that matches the benchmark eye use type.

[0068] Based on the matching results, the data in the second category of eye use data segment are further classified according to the baseline eye use type to generate multiple sets of standardized eye use data groups.

[0069] Randomly select several data points from the first type of eye use data segment as the first analysis sample, and at the same time, randomly select data from each group of standardized eye use data groups as the second analysis sample;

[0070] The first analysis sample is analyzed to determine the gaze feature points of the child's eyes and head. At the same time, a real-time gaze trajectory network of the child is constructed based on the gaze feature points, and the first type of eye use data is deployed to the real-time gaze trajectory network based on the gaze feature points.

[0071] Based on the deployment results, the distribution characteristics of the first type of eye use data in the real-time gaze trajectory network are determined, and the children's real-time gaze deviation habits are determined based on the distribution characteristics.

[0072] A standard gaze reference network is constructed based on the aforementioned baseline eye use type. Simultaneously, the second analysis sample is sequentially input into the standard gaze reference network for posture fitting, and the child's standard gaze reference habits are determined based on the fitting results.

[0073] Obtain the first weight of real-time gaze offset habits and the second weight of standardized gaze reference habits. At the same time, score the real-time gaze offset habits according to the first weight to obtain the first score, and score the standardized gaze reference habits according to the second weight to obtain the second score.

[0074] A comprehensive score for children's eye use is calculated by weighting the first and second scores, and the comprehensive score is compared with a preset scoring threshold to determine whether the child's line of sight deviates from the normal eye use angle.

[0075] When the overall score is greater than or equal to the preset score threshold, the child's line of sight is determined to be normal and their eye posture is correct.

[0076] Otherwise, the system determines that the child's gaze angle is abnormal and deviates from the standard eye use direction, and generates an eye use posture warning when the gaze deviation determination is valid.

[0077] In the above embodiments, the child's eye use behavior monitoring unit acquires target eye use data with data tags, matches it with a preset eye use standard library to determine the baseline eye use type, then classifies the target eye use data to obtain two types of data segments, and then further classifies the second type of data to generate standardized data groups. By extracting and analyzing samples, a real-time gaze trajectory network and a standard gaze reference network are constructed to determine the child's gaze deviation habits and standard reference habits, respectively. Then, a weighted comprehensive score is calculated and compared with a preset threshold to obtain the judgment result. The entire process is based on actual eye use data and combined with an algorithm model, which can accurately identify abnormal postures such as forward head tilt and side tilt, avoiding misjudgment. The judgment method considers both real-time eye use status and references standard habits, making the judgment result more convincing, helping to correct children's poor eye use posture in a timely manner, and reducing the risk of myopia.

[0078] The central processing and control unit has a built-in light environment threshold library for myopia prevention in children and a safe eye behavior standard library, which includes safe eye distance thresholds, maximum continuous eye use duration thresholds, appropriate light intensity ranges, standard color temperature ranges and glare limit values ​​for children of different ages.

[0079] The central processing and control unit has four preset eye-use scenario modes: learning and reading, using electronic devices, reading picture books, and resting and relaxing. It automatically matches the optimal combination of light environment parameters and generates precise light environment adjustment instructions, specifically including:

[0080] It receives and analyzes the target eye use data uploaded by the eye use behavior monitoring unit and the judgment results of whether the line of sight deviates from the normal eye use angle in real time;

[0081] By comparing the target eye use data with the eye use behavior safety standard database, the type of eye use scenario the child is currently in can be determined and the corresponding scenario can be identified.

[0082] Extract the preset target illumination intensity parameters, target color temperature parameters, target color rendering index parameters, and glare suppression parameters for this scene;

[0083] The difference between the current real-time light environment parameters and the target light environment parameters is calculated to obtain the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient.

[0084] Based on the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient, a light environment adjustment command containing the adjustment range, adjustment speed, and adjustment duration is generated and sent to the dynamic light environment adjustment unit for execution.

[0085] This includes determining the type of eye-use scenario the child is currently in and identifying the corresponding scenario, specifically including:

[0086] Acquire target eye use data uploaded by the children's eye use behavior monitoring unit, and extract current light environment parameters synchronously collected by the multi-dimensional light environment perception unit;

[0087] The built-in eye behavior safety standard library is invoked to extract the scene feature parameter set corresponding to the preset scene. The scene feature parameter set includes the standard object-eye distance range, typical eye posture range, characteristic continuous eye use duration threshold and adaptive light environment parameter range for each scene.

[0088] The target eye use data and current light environment parameters are standardized and preprocessed to remove abnormal fluctuation data points and generate a normalized effective data set. Abnormal fluctuation data points are defined as data that exceed the extreme values ​​of the feature parameter sets of each scenario in the eye use behavior safety standard library by 20%.

[0089] Construct a scene matching feature vector by mapping the real-time object-eye distance, continuous eye use duration, real-time eye use posture, gaze angle state, ambient light intensity and color temperature parameters in the effective dataset to vector dimensions, and generate the feature vector to be matched.

[0090] The cosine similarity between the feature vector to be matched and the standard feature vectors corresponding to the four preset scene feature parameter sets is calculated to obtain four similarity coefficients. The similarity coefficients range from 0 to 1, and the closer the value is to 1, the higher the matching degree.

[0091] Set a similarity threshold of 0.75 and filter out preset scenes with similarity coefficients greater than or equal to the threshold as candidate scenes;

[0092] If there is only one candidate scene, the candidate scene will be determined as the type of eye-use scene the child is currently in and locked.

[0093] If there are at least two candidate scenarios, extract the differential feature parameters of the candidate scenarios. The differential feature parameters include screen blue light radiation correlation parameters specific to electronic device usage scenarios, close-range viewing duration parameters specific to picture book viewing scenarios, writing action correlation parameters specific to learning and reading scenarios, and the proportion of time spent looking into the distance specific to rest and relaxation scenarios.

[0094] A secondary matching scoring model is constructed based on differential feature parameters to score each candidate scenario, with each item scored from 0 to 25 points, for a total score of 100 points.

[0095] Substitute the corresponding parameters from the valid dataset into the secondary matching scoring model, calculate the final score for each candidate scenario, and select the candidate scenario with the highest score as the current eye-use scenario type for the child and lock it in.

[0096] If the similarity coefficient of all preset scenes is less than 0.75, the default scene matching mechanism will be activated. Based on the standard eye use characteristics corresponding to the child's current age, the closest preset scene will be matched as the temporary locked scene, and the target eye use data and light environment parameters will be continuously collected in the next 30 seconds for dynamic verification.

[0097] If the similarity coefficient of the temporarily locked scene during the dynamic verification period is consistently greater than or equal to 0.75, the temporarily locked scene will be determined as the final eye-use scene type and locked; if the similarity coefficient is still less than 0.75, the scene abnormality prompt information will be output to the hierarchical warning interaction unit, and the temporarily locked scene will be maintained until the data meets the matching conditions.

[0098] Based on the above embodiments, in order to make the scene determination logic easier to understand, typical cases of different age groups and different eye habits are selected here, and the implementation process is demonstrated step by step according to the determination process.

[0099] Example 1: Determining the viewing scenario of a 6-year-old child's picture book (routine judgment)

[0100] In the eye use behavior safety standard library, preset scenario standards are set as shown in Table 1.

[0101] Scene type Standard eye distance range Typical eye posture Feature continuous eye use duration threshold Suitable for ambient light intensity range Suitable color temperature range Learning to read 30-40cm Sit upright with your eyes perpendicular to the paper. ≤20 minutes 500-700lx 4000-4500K Use of electronic devices 40-60cm Sitting / semi-sitting posture, with eyes slightly tilted downwards. ≤15 minutes 400-600lx 3800-4200K Picture book viewing 25-35cm Sitting or kneeling posture, with your line of sight at a 30-45° angle to the picture book. ≤15 minutes 450-650lx 3900-4300K Rest and relax ≥100cm (distant view) Sitting / standing posture, eyes level or upward. Unrestricted 300-500lx 3500-5000K

[0102] Table 1

[0103] The device synchronously collects target eye data and current light environment parameters: real-time object-eye distance is 32cm; continuous eye use duration is 8 minutes; real-time eye use posture is kneeling, with the line of sight forming a 38° angle with the picture book; the line of sight angle determination result is no deviation; ambient light intensity is 520lx; color temperature is 4100K; illuminance distribution parameter is uniformity ≥0.85;

[0104] The measured data did not exceed 20% of the extreme values ​​of the feature parameter sets for each scenario, thus generating a valid dataset;

[0105] Constructing feature vectors: Generate feature vectors to be matched = [32cm, 8min, kneeling posture at 38° angle, no deviation, 520lx, 4100K];

[0106] Cosine similarity was calculated: similarity to the standard vector of the picture book viewing scenario: 0.92; similarity to the standard vector of the learning and reading scenario: 0.68; similarity to the standard vector of the electronic device usage scenario: 0.55; similarity to the standard vector of the rest and relaxation scenario: 0.31.

[0107] Candidate scene filtering: Scenes with a similarity of ≥0.75 are only "picture book viewing" and are directly locked;

[0108] The current scene is determined to be "picture book viewing". The central processing control unit calls the optimal light environment parameters for this scene (light intensity 550lx, color temperature 4200K, glare suppression parameter ≤10) to perform adjustment.

[0109] Example 2: Determining the electronic device usage scenarios of a 10-year-old child (multiple candidate scenarios, secondary matching)

[0110] Core differentiation parameters are set for the preset scenario standards. The core differentiation parameters used by electronic devices are: screen blue light radiation correlation parameter ≥ 0.6, and object-eye distance 40-60cm. The core differentiation parameters for learning and reading are: writing action correlation parameter ≥ 0.7, and object-eye distance 30-40cm.

[0111] The device synchronously collects target eye data and current light environment parameters: real-time object-eye distance: 45cm; continuous eye use duration: 12 minutes; real-time eye posture: semi-sitting posture, with the line of sight tilted slightly downward at 15°; line of sight angle determination result: no deviation; blue light radiation correlation parameter: 0.65; writing action correlation parameter: 0.3; ambient light intensity: 480lx; color temperature: 4000K;

[0112] Constructing feature vectors: Generate a feature vector to be matched = [45cm, 12min, half-sitting posture 15° tilt, no deviation, 480lx, 4000K, blue light 0.65, writing 0.3];

[0113] Cosine similarity was calculated: similarity for electronic device usage scenario: 0.83; similarity for learning and reading scenario: 0.76; similarity for other scenarios was <0.75, generating candidate scenarios: electronic device usage, learning and reading;

[0114] Table 2 shows the secondary matching and scoring for candidate scenarios;

[0115] Blue light radiation correlation parameters 25 (full value, 0.65 ≥ 0.6) 5 (Not up to standard) Writing action associated parameters 5 (Not up to standard) 20 (0.3 < 0.7, deduct points) Object-eye distance matching 24 (45cm is between 40-60cm, close to the median) 18 (45cm exceeds the 30-40cm range) Eye posture matching 25 (Semi-seated, tilted posture) 22 (Stricter posture requirements, slight deduction) Light environment parameter adaptability 21 (480lx, 4000K compatible) 15 (Light intensity slightly lower than the 500 lx standard) Total Score 99 points 60 points

[0116] Table 2

[0117] Among them, the electronic device usage scenario scored the highest (99 points). Lock the scenario, adjust the light environment to a light intensity of 500 lx and a color temperature of 4100 K, and activate the blue light filter to assist in the adjustment.

[0118] Example 3: Determining Rest and Relaxation Scenarios for 8-Year-Old Children (Low Similarity Dynamic Validation)

[0119] The device synchronously collects target eye data and current light environment parameters: real-time object-eye distance: 95cm; continuous eye use duration: 5 minutes; real-time eye use posture: standing, line of sight horizontal; line of sight angle determination result: no deviation; ambient light intensity: 320lx; color temperature: 4800K.

[0120] Cosine similarity was calculated: similarity for the rest and relaxation scene was 0.73 (slightly below the 0.75 threshold); similarity for other scenes was <0.6, and there were no candidate scenes that met the threshold; based on the default scene matching of standard eye use characteristics of an 8-year-old child, the "rest and relaxation" scene was temporarily locked;

[0121] Actual data measured after 30 seconds: object-eye distance 110cm; continuous eye use time 5.5 minutes; posture maintained at a standing position with horizontal line of sight; light intensity 330lx; color temperature 4700K;

[0122] Similarity recalculated: the similarity of the rest and relaxation scene improved to 0.81 (≥0.75);

[0123] Dynamic verification passed, confirming the locking of the "Rest and Relax" scene, keeping the light environment parameters (330lx, 4700K) unchanged, and triggering the "Remind me to look into the distance every 10 minutes" function.

[0124] In the above embodiments, the optimal combination of light environment parameters is automatically matched through an efficient process to generate precise adjustment instructions, ensuring that the light environment is suitable for children's eye use scenarios. Two types of standard libraries are built-in, with four preset eye use scenario modes. Relevant data is received and parsed, and the scenario type is initially determined by comparing it with the standard libraries. Candidate scenarios are then filtered by constructing feature vectors and calculating cosine similarity. A secondary matching scoring model is used to determine the final scenario from multiple candidate scenarios, and a dynamic verification mechanism is activated when the similarity is low.

[0125] Once the scene is determined, the target light environment parameters are extracted and calculated with the current parameters to obtain the adjustment coefficient. This generates an instruction containing the adjustment range, speed, and duration. This scene determination method is accurate and efficient, capable of handling various complex situations and ensuring accurate scene locking. Based on precise scene matching, the light environment adjustment makes parameters such as light intensity and color temperature fit the scene requirements. For example, the electronic device usage scene will be adapted to appropriate light intensity and color temperature, and blue light filtering will also be activated. Through dynamic light environment adjustment, children can obtain comfortable and safe light in different eye use scenarios, reducing eye burden and helping to prevent and control myopia.

[0126] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A device for the prevention and control of myopia in children based on light environment regulation, characterized in that, It includes the main body of the device, a central processing and control unit, a multi-dimensional light environment sensing unit, a dynamic light environment adjustment unit, a children's eye use behavior monitoring unit, a graded early warning and interaction unit, a data storage unit, and a power supply unit; The multi-dimensional light environment sensing unit is used to collect light environment parameters in children's learning and living spaces in real time; The child eye behavior monitoring unit is used to acquire the child's target eye data in a non-contact manner, determine whether the line of sight deviates from the normal eye angle, and generate eye posture warning information. The central processing and control unit is used to perform fusion analysis and logical judgment on the collected light environment parameters and eye behavior data, generate adaptive adjustment instructions according to the optical standards for myopia prevention in children, drive the dynamic light environment adjustment unit to output appropriate lighting light, and optimize the local eye light environment in real time. The graded early warning interaction unit is used to issue graded reminders when the light environment is not up to standard or when eye use behavior is abnormal. The data storage unit is used to record control data and eye habits; The power supply unit is used to provide operating power.

2. The children's myopia prevention and control device based on light environment regulation as described in claim 1, characterized in that, The multi-dimensional light environment sensing unit includes a high-precision light intensity sensor, a color temperature acquisition sensor, a blue light sensor, a color rendering index detection module, a glare detection module, and an illuminance uniformity analysis module. The multi-dimensional light environment sensing unit collects light environment parameters and transmits them to the central processing and control unit through a time-division synchronous acquisition mechanism.

3. The children's myopia prevention and control device based on light environment regulation as described in claim 1, characterized in that, The child eye behavior monitoring unit combines an infrared ranging module and a high-definition image recognition module. It uses a binocular visual positioning algorithm to monitor the real-time object-eye distance between the child's eyes and the target object, while recording the duration of continuous eye use, identifying the real-time eye posture and determining whether it is an abnormal posture, and judging whether the line of sight deviates from the normal eye use angle.

4. The children's myopia prevention and control device based on light environment regulation as described in claim 3, characterized in that, The child's eye behavior monitoring unit determines whether the line of sight deviates from the normal eye angle, specifically including: Based on the child eye use behavior monitoring unit, the target eye use data for monitoring the child's eyes is determined. The target eye use data includes real-time object-eye distance, continuous eye use duration and real-time eye use posture. At the same time, the data tags of the target eye use data are obtained. The data tags of the target eye use data are input into a preset eye use standard library for matching, and the benchmark eye use type for standardizing the judgment of children's eye posture is output based on the matching results. The preset eye use standard library includes benchmark eye use types for looking straight ahead, reading, writing, looking into the distance, and watching screens. The target eye use data is matched item by item with the benchmark eye use type, and the target eye use data is initially classified based on the matching results to generate a first type of eye use data segment and a second type of eye use data segment. The first type of eye use data segment consists of the target eye use data that does not match the benchmark eye use type, and the second type of eye use data segment consists of the target eye use data that matches the benchmark eye use type. Based on the matching results, the data in the second category of eye use data segment are further classified according to the baseline eye use type to generate multiple sets of standardized eye use data groups. Randomly select several data points from the first type of eye use data segment as the first analysis sample, and at the same time, randomly select data from each group of standardized eye use data groups as the second analysis sample; The first analysis sample is analyzed to determine the gaze feature points of the child's eyes and head. At the same time, a real-time gaze trajectory network of the child is constructed based on the gaze feature points, and the first type of eye use data is deployed to the real-time gaze trajectory network based on the gaze feature points. Based on the deployment results, the distribution characteristics of the first type of eye use data in the real-time gaze trajectory network are determined, and the children's real-time gaze deviation habits are determined based on the distribution characteristics. A standard gaze reference network is constructed based on the benchmark eye use type. At the same time, the second analysis sample is sequentially input into the standard gaze reference network for posture fitting, and the child's standard gaze reference habit is determined based on the fitting results. Obtain the first weight of real-time gaze offset habits and the second weight of standardized gaze reference habits. At the same time, score the real-time gaze offset habits according to the first weight to obtain the first score, and score the standardized gaze reference habits according to the second weight to obtain the second score. A comprehensive score for children's eye use is calculated by weighting the first and second scores, and the comprehensive score is compared with a preset scoring threshold to determine whether the child's line of sight deviates from the normal eye use angle. When the overall score is greater than or equal to the preset score threshold, the child's line of sight is determined to be normal and their eye posture is correct. Otherwise, the system determines that the child's gaze angle is abnormal and deviates from the standard eye use direction, and generates an eye use posture warning when the gaze deviation determination is valid.

5. The child myopia prevention and control device based on light environment regulation as described in claim 1, characterized in that, The central processing and control unit has a built-in threshold library for light environment for myopia prevention in children and a standard library for safe eye use behavior, which includes safe eye use distance thresholds, maximum continuous eye use duration thresholds, appropriate light intensity ranges, standard color temperature ranges and glare limit values ​​for children of different age groups. The central processing and control unit has four preset eye-use scenario modes: learning and reading, using electronic devices, reading picture books, and resting and relaxing. It automatically matches the optimal combination of light environment parameters and generates precise light environment adjustment instructions, specifically including: It receives and analyzes the target eye use data uploaded by the eye use behavior monitoring unit and the judgment results of whether the line of sight deviates from the normal eye use angle in real time; By comparing the target eye use data with the eye use behavior safety standard database, the type of eye use scenario the child is currently in can be determined and the corresponding scenario can be identified. Extract the preset target illumination intensity parameters, target color temperature parameters, target color rendering index parameters, and glare suppression parameters for this scene; The difference between the current real-time light environment parameters and the target light environment parameters is calculated to obtain the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient. Based on the brightness adjustment coefficient, color temperature adjustment coefficient, and light diffusion coefficient, a light environment adjustment command containing the adjustment range, adjustment speed, and adjustment duration is generated and sent to the dynamic light environment adjustment unit for execution.

6. The child myopia prevention and control device based on light environment regulation as described in claim 5, characterized in that, Determine the type of eye-use scenario the child is currently in and pinpoint the corresponding scenario, specifically including: Acquire target eye use data uploaded by the children's eye use behavior monitoring unit, and extract current light environment parameters synchronously collected by the multi-dimensional light environment perception unit; The built-in eye behavior safety standard library is invoked to extract the scene feature parameter set corresponding to the preset scene. The scene feature parameter set includes the standard object-eye distance range, typical eye posture range, characteristic continuous eye use duration threshold and adaptive light environment parameter range for each scene. Standardize and preprocess the target eye use data and current light environment parameters, remove abnormal fluctuation data points, and generate a normalized effective data set. Construct a scene matching feature vector by mapping the real-time object-eye distance, continuous eye use duration, real-time eye use posture, gaze angle state, ambient light intensity and color temperature parameters in the effective dataset to vector dimensions, and generate the feature vector to be matched. Calculate the cosine similarity between the feature vector to be matched and the standard feature vectors corresponding to four preset scene feature parameter sets to obtain four similarity coefficients, the value of which ranges from 0 to 1; Set a similarity threshold of 0.75, filter out preset scenes with similarity coefficients greater than or equal to the threshold as candidate scenes, and then filter and lock the final eye-use scene type from the candidate scenes.

7. The child myopia prevention and control device based on light environment regulation as described in claim 6, characterized in that, The final eye-use scenario type is selected and locked from the candidate scenarios, specifically including: After filtering out candidate scenarios, if there is only one candidate scenario, the candidate scenario is determined to be the type of eye-use scenario in which the child is currently experiencing and is locked. If there are at least two candidate scenarios, extract the differential feature parameters of the candidate scenarios. The differential feature parameters include screen blue light radiation correlation parameters specific to electronic device usage scenarios, close-range viewing duration parameters specific to picture book viewing scenarios, writing action correlation parameters specific to learning and reading scenarios, and the proportion of time spent looking into the distance specific to rest and relaxation scenarios. A secondary matching scoring model is constructed based on differential feature parameters to score each candidate scenario, with each item scored from 0 to 25 points, for a total score of 100 points. Substitute the corresponding parameters from the valid dataset into the secondary matching scoring model, calculate the final score for each candidate scenario, and select the candidate scenario with the highest score as the current eye-use scenario type for the child and lock it in. If the similarity coefficient of all preset scenes is less than 0.75, the default scene matching mechanism will be activated. Based on the standard eye use characteristics corresponding to the child's current age, the closest preset scene will be matched as the temporary locked scene, and the target eye use data and light environment parameters will be continuously collected in the next 30 seconds for dynamic verification. If the similarity coefficient of the temporarily locked scene during the dynamic verification period is consistently greater than or equal to 0.75, the temporarily locked scene will be determined as the final eye-use scene type and locked; if the similarity coefficient is still less than 0.75, the scene abnormality prompt information will be output to the hierarchical warning interaction unit, and the temporarily locked scene will be maintained until the data meets the matching conditions.

8. The child myopia prevention and control device based on light environment regulation as described in claim 5, characterized in that, The dynamic light environment adjustment unit includes a full-spectrum LED light source array, a PWM pulse width adjustment module, a color temperature continuous adjustment module, a glare suppression module, and a light diffusion module. The dynamic light environment adjustment unit is mounted on the main body of the device.

9. The child myopia prevention and control device based on light environment regulation as described in claim 8, characterized in that, The main body of the device is equipped with a multi-functional installation and fixing unit, which includes three installation structures: a clamping base, a wall-mounted bracket, and a desktop support.

10. The child myopia prevention and control device based on light environment regulation as described in claim 1, characterized in that, The graded early warning interaction unit includes a voice prompt module, an LED status indicator module, a vibration reminder module, and a text display module. The graded early warning is divided into three levels of response based on the risk level: mild reminder, moderate warning, and severe alarm. At the same time, abnormal information, eye posture warning information, and control records are pushed to the parent terminal device simultaneously via wireless communication.

Images