Myopia out-of-focus protection vision training method based on free-form surface optics

Abstract

The invention provides a myopia out-of-focus protection visual training method based on free-form surface optics, and relates to the technical field of visual protection. The shortsightedness out-of-focus protection vision training method based on free-form surface optics comprises the following specific steps: S1, equipment construction: configuring an out-of-focus read-write table with a free-form surface optical system, and installing a high-frame camera at a proper position of the out-of-focus read-write table for capturing images of two eyes of a user; a user database is established during first use, age, vision and binocular information is input, the user state is monitored in real time in the use process, the virtual imaging distance is automatically set according to the formula delta R = 0.2 * D * A, and meanwhile, the light path parameters of the free-form surface optical system are adjusted. According to the dynamic adjustment mode based on the user personalized data, the vision conditions and age characteristics of different users can be accurately matched, the problem of insufficient suitability in the prior art is solved, and the pertinence and effectiveness of vision training are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vision protection, and in particular to a myopia defocus protection vision training method based on free-form surface optics. Background Art

[0002] The Defocus Reading and Writing Desk's unique AR free-form surface design allows for zooming out of a nearby desk book to a distance of 3-8 meters, transforming close-up vision into distant viewing. This allows the eyes to remain relaxed while looking at distant objects, reducing the risk of accommodation problems and even myopia caused by over-accommodation due to close-up vision. The Defocus Reading and Writing Desk also provides ample reading light, facilitating adjustments to poor sitting posture and reducing the risk of myopia caused by poor eye habits.

[0003] The existing overpass reading and writing stations use the user's self-perception to adjust different distant image distances (3m-8m), which can theoretically alleviate the adjustment load caused by close-range eye use, but there are still significant technical bottlenecks: First, it relies on manual adjustment of the imaging distance and cannot dynamically optimize parameters based on the user's vision, age and real-time fatigue status, resulting in insufficient adaptability; second, due to the lack of multimodal biometric monitoring (such as pupil dynamics, eyelid movement, etc.), it is difficult to accurately identify early visual fatigue; in addition, the fixed imaging distance easily causes the ciliary muscle to produce adaptive adjustment, and long-term use may weaken the prevention and control effect due to the single stimulation; most importantly, the existing system lacks a quantitative evaluation and feedback mechanism, and cannot form a personalized closed-loop intervention of "monitoring-training-optimization", which restricts the accuracy and effectiveness of myopia prevention and control.

[0004] To this end, we have developed a new myopia defocus protection visual training method based on free-form surface optics. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a myopia defocus protection visual training method based on free-form surface optics, which solves the problem that the existing defocus reading and writing tables lack dynamic adaptation, precise fatigue monitoring, diversified training modes and quantitative evaluation systems, resulting in unstable myopia prevention and control effects and insufficient personalization.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a myopia defocus protection vision training method based on free-form surface optics, comprising the following specific steps:

[0009] S1. Equipment Setup: Configure a defocused reading and writing table with a free-form optical system. Install a high-frame camera at a suitable location on the defocused reading and writing table to capture images from the user's eyes. Also place an electrically adjustable book placement table, a voice prompter, and an industrial control all-in-one computer at a suitable location.

[0010] S2. Control connection: Electrically connect the built-in chip of the industrial control all-in-one computer with the control module of the defocused reading and writing table, the image processing module of the high-frame camera, the electric control module of the electrically liftable book placement table, and the voice prompt control module to achieve real-time data transmission and device linkage;

[0011] S3. Parameter setting: When using for the first time, a user database is established and the user's age, vision, and binocular information are entered. The binocular information is analyzed through 3 minutes of natural eye movement to obtain initial pupil changes, blink frequency, and facial expression data;

[0012] S4. Training settings: Automatically set the virtual imaging distance based on user parameters and adjust the optical path parameters of the free-form surface optical system to achieve the target imaging distance. For example, for a 10-year-old user with 100-degree myopia, the initial virtual imaging distance is set to 5 meters, and the optical path parameters of the free-form surface optical system are adjusted according to the formula ΔR = 0.2 × D × A to achieve imaging at this distance;

[0013] S5, Real-time monitoring: Continuously collect dynamic image data of both eyes through a high-frame camera while reading;

[0014] S6. Fatigue judgment: Analyze real-time eye movement trajectory, pupil diameter change rate, blink frequency abnormality and facial muscle tension. When at least three of the four indicators exceed the preset threshold, it is judged as visual fatigue.

[0015] S7. Start training: When fatigue is detected for a period exceeding a preset time, the training mode is activated. The electric lifting platform and the free-form optical system are used to shorten the virtual imaging distance, and voice-guided eye movement training is performed simultaneously.

[0016] S8. Data evaluation: After the training is completed, the system compares the user's pupil response and blink frequency data before and after the training, and calculates the pupil recovery index and blink frequency improvement rate;

[0017] S9. Optimization and adjustment: If the fatigue relief index does not meet expectations, the training parameters will be dynamically adjusted through the reinforcement learning algorithm and training will be carried out again.

[0018] Preferably, the free-form surface optical system in S1 includes:

[0019] The curvature of the aspheric lens group is adjustable, and the curvature adjustment range is 200-1000mm;

[0020] Dynamic diffractive optical element, using 32-order phase-type liquid crystal diffractive optical element, can achieve continuous adjustment of virtual imaging distance within the range of 0.5-10m;

[0021] Polarization splitter film layer, with a transmittance of >92%@530nm, is used to separate the display light path from the imaging light path.

[0022] Preferably, the S1 high frame camera is equipped with an 850nm infrared fill light module with an irradiance of less than 0.8mW / cm 2 , it can capture clear eye images in low-light environments. The electric liftable book placement table in the S1 has a lifting stroke of 5-15cm, a lifting accuracy of ±0.5mm, a response time of less than 0.5 seconds, and an adjustable lifting speed range of 1-5cm / s.

[0023] Preferably, in the formula of S4, D is the degree of myopia, and A is the age coefficient (A=0.02×(14-age) when age<14; A=0 when 14≤age≤18; A=0.01×(age-18) when age>18).

[0024] Preferably, in S5, the eye movement trajectory is recorded with an accuracy of 0.1 mm using the StarBurst algorithm;

[0025] Calculate pupil diameter change curve through iris recognition algorithm;

[0026] Facial micro-expression recognition based on convolutional neural networks;

[0027] A 3D eyelid motion model containing 200 feature points is established.

[0028] Preferably, the fatigue judgment standard in S6 is:

[0029] Blinking frequency is more than 22 times / minute or less than 8 times / minute;

[0030] The pupil diameter fluctuation is greater than 35% of the initial value;

[0031] The cumulative gaze deviation exceeds 15°;

[0032] The EMG signal intensity of the glabellar muscle reached more than 2 times that of the resting state.

[0033] Preferably, the optical path parameter adjustment in 7 includes:

[0034] The curvature of the aspheric lens is adjusted according to the formula ΔR = 0.2 × D × A, where D is the degree of myopia and A is the age coefficient;

[0035] The phase distribution of the diffraction element is updated in real time according to the virtual image distance;

[0036] The refraction angle of the optical path matches the lifting displacement of the book placement table at a ratio of 1:1.2.

[0037] Preferably, the training in S7 includes:

[0038] Phase 1: The book is lowered by 3-5 cm at a speed of 2 cm / s using an electric lifting platform. The free-form optical system is simultaneously adjusted to shorten the virtual imaging distance by 30-50%, which lasts for 1 minute.

[0039] The second stage: Generate a dynamic training mode in which a 3-meter clear image and an 8-meter blurred image are displayed alternately at a frequency of 0.3-3 Hz;

[0040] Voice guidance: "Look up, hold for 5 seconds... look down, blink 3 times..." Repeat eye movement training.

[0041] Preferably, the evaluation indicators in S8 include:

[0042] Pupil recovery index: the time required for pupil diameter to return to baseline value after training;

[0043] Blink frequency improvement rate: (post-training frequency - pre-training frequency) / pre-training frequency × 100%;

[0044] Eye movement activity value: the ratio of the complexity of eye movement trajectories after training to the baseline value.

[0045] Preferably, the S9 further includes:

[0046] Record the effect data of each training session to form a personal adaptation curve;

[0047] Use reinforcement learning algorithm to optimize the next training parameters;

[0048] When the training efficiency is lower than 70% for three consecutive times, the system parameter calibration is triggered.

[0049] (3) Beneficial effects

[0050] The present invention provides a myopia defocus protection vision training method based on free-form surface optics. It has the following beneficial effects:

[0051] 1. This free-form surface optics-based myopia defocus protection visual training method establishes a user database upon first use, recording age, vision, and binocular information. It then monitors the user's status in real time during use, automatically setting the virtual imaging distance based on the formula ΔR = 0.2 × D × A, and simultaneously adjusting the optical path parameters of the free-form surface optical system. This dynamic adjustment method, based on personalized user data, accurately matches the vision and age characteristics of different users, overcoming the limited adaptability of existing technologies and significantly improving the targetedness and effectiveness of visual training.

[0052] 2. This free-form optics-based visual training method for myopia defocus protection uses a high-frame camera integrated with an algorithm to monitor multiple indicators in real time, including pupil diameter fluctuation, abnormal blink rate, eye movement trajectory deviation, and glabellar muscle tension. Fatigue detection is triggered when at least three of these indicators exceed thresholds. Capturing subtle changes such as pupil fluctuation or abnormal blink rate overcomes the limitations of existing technologies that lack biometric monitoring and hinder early detection of visual fatigue, providing a reliable basis for subsequent intervention.

[0053] 3. This myopia defocus protection visual training method based on free-form surface optics, after the training, the system compares the user's pupil reaction and blink frequency data before and after training, calculates evaluation indicators such as pupil recovery index, blink frequency improvement rate and eye movement vitality value, and records the effect data of each training to form a personal adaptation curve. If the fatigue relief index does not meet expectations, the reinforcement learning algorithm is used to optimize the parameters of the next training. When the efficiency of three consecutive training sessions is less than 70%, the system parameter calibration is triggered, thus building a complete closed loop of "monitoring-training-optimization". This personalized dynamic optimization mechanism effectively avoids the problem of adaptive adjustment of the ciliary muscle caused by fixed imaging distance, continuously optimizes the training plan, and significantly enhances the accuracy and effectiveness of myopia prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] like Figure 1 As shown, an embodiment of the present invention provides a myopia defocus protection vision training method based on free-form surface optics, comprising the following specific steps:

[0057] S1. Equipment Setup

[0058] Before implementing this training method, equipment must be set up. A defocused reading and writing platform with a free-form optical system is configured. This free-form optical system includes an aspheric lens group with adjustable curvature, a dynamic diffractive optical element, and a polarization beam splitter film layer. The aspheric lens group with adjustable curvature has a curvature adjustment range of 200-1000mm and can be flexibly adjusted according to imaging requirements. The dynamic diffractive optical element uses a 32-order phase-type liquid crystal diffractive optical element, which can achieve continuous adjustment of the virtual imaging distance within the range of 0.5-10m. The polarization beam splitter film layer has a transmittance of >92% @ 530nm, effectively separating the display light path from the imaging light path.

[0059] A high-frame camera equipped with an 850nm infrared fill light module is installed at a suitable position on the defocused reading and writing platform. The irradiance of the infrared fill light module is less than 0.8mW / cm 2 , ensuring clear eye images in low-light environments. Meanwhile, a power-operated book stand, voice prompter, and industrial control computer are conveniently located. The power-operated book stand has a travel range of 5-15cm, a lifting accuracy of ±0.5mm, a response time of less than 0.5 seconds, and an adjustable lifting speed range of 1-5cm / s, meeting the height adjustment requirements for different training scenarios.

[0060] S2. Control connection

[0061] The built-in chip of the industrial control all-in-one computer is electrically connected to the control module of the defocused reading and writing table, the image processing module of the high-frame camera, the electric control module of the electrically liftable book placement table, and the voice prompt control module. This connection uses a high-speed data transmission protocol to build a stable data interaction channel to ensure zero-delay transmission of instructions and data between modules. The eye image data captured by the high-frame camera is analyzed by the image processing module and can be transmitted to the industrial control all-in-one computer in milliseconds. Its built-in algorithm synchronously analyzes the user's visual fatigue status; at the same time, the defocused reading and writing table control module provides real-time feedback on the parameters of the free-form surface optical system. When user fatigue is detected, the industrial control all-in-one computer immediately sends a lifting command to the electric control module of the book placement table, and synchronously adjusts the optical path parameters of the optical system, and cooperates with the voice prompt to start guided training, realizing seamless coordination of the entire link from data acquisition and analysis to equipment response, ensuring stable and efficient operation of the entire training system.

[0062] S3. Parameter settings

[0063] When a user uses this training system for the first time, the system will establish a user database. The user's age and vision information will be obtained through manual entry. At the same time, a high-frame camera will be used to conduct eye movement analysis in a natural state for 3 minutes to collect information about the user's eyes, including initial pupil changes, blinking frequency, and facial expression data. These data will serve as basic parameters for subsequent training settings and fatigue judgment. For example, for a 12-year-old user with 150 degrees of myopia, after entering age and vision information, the initial pupil diameter, blinking frequency and other data will be obtained through eye movement analysis and stored in the user database.

[0064] S4. Training settings

[0065] According to the age, vision and other parameters entered by the user, the system automatically sets the virtual imaging distance and adjusts the optical path parameters of the free-form optical system to achieve the target imaging distance. Specifically, the calculation and adjustment are based on the formula ΔR = 0.2 × D × A (where D is the degree of myopia, A is the age coefficient, A = 0.02 × (14-age) when age < 14; A = 0 when 14 ≤ age ≤ 18; A = 0.01 × (age-18) when age > 18). For example, for the above-mentioned 12-year-old user with 150 degrees of myopia, the age coefficient A = 0, and according to the formula, ΔR = 0 can be obtained. Assuming that the virtual imaging distance initially set by the system is 6 meters, the target imaging distance is achieved by adjusting the optical path parameters of the free-form optical system.

[0066] S5. Real-time monitoring

[0067] While the user is reading, a high-frame camera continuously captures dynamic image data from both eyes. Using the StarBurst algorithm, it records eye movement with 0.1mm accuracy. An iris recognition algorithm calculates pupil diameter curves. A convolutional neural network is used to identify facial micro-expressions. A three-dimensional eyelid motion model containing 200 feature points is built to comprehensively and accurately monitor the user's eye status. The collected data is transmitted in real time to an industrial control computer for processing and analysis.

[0068] S6. Fatigue judgment

[0069] The industrial computer analyzes real-time data collected from eye movements, pupil diameter change rate, blink rate anomalies, and facial muscle tension. If at least three of these four indicators—blink rate exceeding 22 times / minute or falling below 8 times / minute, pupil diameter fluctuation exceeding 35% of the initial value, cumulative gaze deviation exceeding 15°, or glabellar muscle EMG signal strength exceeding twice the resting state—exceed preset thresholds, the user is considered to be experiencing visual fatigue. For example, if during the monitoring process, the user's blinking frequency is 25 times / minute, the pupil diameter fluctuation amplitude is 40% of the initial value, the cumulative gaze offset is 12°, and the interocular muscle EMG signal strength is 1.8 times that of the resting state, the blinking frequency and pupil diameter fluctuation amplitude exceed the threshold and do not meet the judgment conditions; if the blinking frequency is 26 times / minute, the pupil diameter fluctuation amplitude is 45% of the initial value, the cumulative gaze offset is 16°, and the interocular muscle EMG signal strength is 2.2 times that of the resting state, then three of the four indicators exceed the threshold, and the user is judged to be in a state of visual fatigue.

[0070] S7. Start training

[0071] When the system continuously detects that the user is in a state of fatigue and this continues for more than a preset time, the training mode is activated. First, the height of the book is lowered by 3-5cm at a speed of 2cm / s using an electric lifting platform. At the same time, the free-form optical system is adjusted synchronously to shorten the virtual imaging distance by 30-50%. This process lasts for 1 minute. Then, the system enters the second stage, generating a dynamic training mode in which a 3-meter clear image and an 8-meter blurred image are displayed alternately at a frequency of 0.3-3Hz. During this process, the voice prompter performs eye movement training in a cycle according to the guidance content of "Look up, hold for 5 seconds... Look down, blink 3 times..." to help users relieve visual fatigue.

[0072] S8. Data Evaluation

[0073] After training, the system compares the user's pupillary response and blink rate data before and after training, calculating evaluation metrics such as the pupil recovery index (i.e., the time required for pupil diameter to return to baseline value after training), blink rate improvement rate (calculated as (post-training frequency - pre-training frequency) / pre-training frequency × 100%), and eye movement vitality value (the ratio of post-training eye movement trajectory complexity to baseline value). These metrics quantify the training effect and provide data support for subsequent optimization and adjustment.

[0074] S9. Optimization and Adjustment

[0075] If fatigue relief indicators fail to meet expectations, the system records the effectiveness of each training session to create a personalized adaptation curve and uses a reinforcement learning algorithm to optimize parameters for the next training session. For example, adjustments may be made to parameters such as virtual imaging distance, book table height, and image display frequency. If the effectiveness rate for three consecutive training sessions falls below 70%, system parameter calibration is triggered, and user parameters are re-set and evaluated to ensure that the training plan consistently meets the user's actual needs and achieves optimal myopia defocus protection and visual training results.

[0076] Examples of parameter calculation for multiple sets of data. The results are shown in the table below:

[0077]

[0078] Examples of multiple sets of data for fatigue determination, the results are shown in the table below:

[0079]

[0080] The training results and evaluation of multiple sets of data are shown in the following table:

[0081]

[0082]

[0083] In summary, the present invention realizes precise adaptation, scientific monitoring and efficient intervention of myopia defocus protection visual training based on dynamic optical path adjustment, multi-dimensional visual fatigue monitoring, two-stage linkage training and quantitative evaluation optimization mechanism based on user age, vision and other parameters, and significantly improves the myopia prevention and control effect.

[0084] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A myopia defocus protection vision training method based on free-form surface optics, characterized in that: The specific steps include: S1. Equipment Setup: Configure a defocused reading and writing table with a free-form optical system. Install a high-frame camera at a suitable location on the defocused reading and writing table to capture images from the user's eyes. Also place an electrically adjustable book placement table, a voice prompter, and an industrial control all-in-one computer at a suitable location. S2. Control connection: Electrically connect the built-in chip of the industrial control all-in-one computer with the control module of the defocused reading and writing table, the image processing module of the high-frame camera, the electric control module of the electrically liftable book placement table, and the voice prompt control module to achieve real-time data transmission and device linkage; S3. Parameter setting: When using for the first time, a user database is established and the user's age, vision, and binocular information are entered. The binocular information is analyzed through 3 minutes of natural eye movement to obtain initial pupil changes, blink frequency, and facial expression data; S4. Training settings: Automatically set the virtual imaging distance based on user parameters and adjust the optical path parameters of the free-form surface optical system to achieve the target imaging distance. For example, for a 10-year-old user with 100-degree myopia, the initial virtual imaging distance is set to 5 meters, and the optical path parameters of the free-form surface optical system are adjusted according to the formula ΔR = 0.2 × D × A to achieve imaging at this distance; S5, Real-time monitoring: Continuously collect dynamic image data of both eyes through a high-frame camera while reading; S6. Fatigue judgment: Analyze real-time eye movement trajectory, pupil diameter change rate, blink frequency abnormality and facial muscle tension. When at least three of the four indicators exceed the preset threshold, it is judged as visual fatigue. S7. Start training: When fatigue is detected for a period exceeding a preset time, the training mode is activated. The electric lifting platform and the free-form optical system are used to shorten the virtual imaging distance, and voice-guided eye movement training is performed simultaneously. S8. Data evaluation: After the training is completed, the system compares the user's pupil response and blink frequency data before and after the training, and calculates the pupil recovery index and blink frequency improvement rate; S9. Optimization and adjustment: If the fatigue relief index does not meet expectations, the training parameters will be dynamically adjusted through the reinforcement learning algorithm and training will be carried out again.

2. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The free-form surface optical system in S1 includes: The curvature of the aspheric lens group is adjustable, and the curvature adjustment range is 200-1000mm; Dynamic diffractive optical element, using 32-order phase-type liquid crystal diffractive optical element, can achieve continuous adjustment of virtual imaging distance within the range of 0.5-10m; Polarization splitter film layer, with a transmittance of >92%@530nm, is used to separate the display light path from the imaging light path.

3. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The S1 medium and high frame camera is equipped with an 850nm infrared fill light module with an irradiance of less than 0.8mW / cm 2 , it can capture clear eye images in low-light environments. The electric liftable book placement table in the S1 has a lifting stroke of 5-15cm, a lifting accuracy of ±0.5mm, a response time of less than 0.5 seconds, and an adjustable lifting speed range of 1-5cm / s.

4. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: In the formula of S4, D is the degree of myopia, and A is the age coefficient (A=0.02×(14-age) when age<14; A=0 when 14≤age≤18; A=0.01×(age-18) when age>18).

5. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: In the S5, the eye movement trajectory is recorded with an accuracy of 0.1 mm using the StarBurst algorithm; Calculate pupil diameter change curve through iris recognition algorithm; Facial micro-expression recognition based on convolutional neural networks; A 3D eyelid motion model containing 200 feature points is established.

6. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The fatigue judgment criteria in S6 are: Blinking frequency is more than 22 times / minute or less than 8 times / minute; The pupil diameter fluctuation is greater than 35% of the initial value; The cumulative gaze deviation exceeds 15°; The EMG signal intensity of the glabellar muscle reached more than 2 times that of the resting state.

7. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The optical path parameter adjustment in 7 includes: The curvature of the aspheric lens is adjusted according to the formula ΔR = 0.2 × D × A, where D is the degree of myopia and A is the age coefficient; The phase distribution of the diffraction element is updated in real time according to the virtual image distance; The refraction angle of the optical path matches the lifting displacement of the book placement table at a ratio of 1:1.

2.

8. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The training in S7 includes: Phase 1: The book is lowered by 3-5 cm at a speed of 2 cm / s using an electric lifting platform. The free-form optical system is simultaneously adjusted to shorten the virtual imaging distance by 30-50%, which lasts for 1 minute. The second stage: Generate a dynamic training mode in which a 3-meter clear image and an 8-meter blurred image are displayed alternately at a frequency of 0.3-3 Hz; Voice guidance: "Look up, hold for 5 seconds... look down, blink 3 times..." Repeat eye movement training.

9. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The evaluation indicators in S8 include: Pupil recovery index: the time required for pupil diameter to return to baseline value after training; Blink frequency improvement rate: (post-training frequency - pre-training frequency) / pre-training frequency × 100%; Eye movement activity value: the ratio of the complexity of eye movement trajectories after training to the baseline value.

10. The myopia defocus protection vision training method based on free-form surface optics according to claim 1, characterized in that: The S9 also includes: Record the effect data of each training session to form a personal adaptation curve; Use reinforcement learning algorithm to optimize the next training parameters; When the training efficiency is lower than 70% for three consecutive times, the system parameter calibration is triggered.

Images