Intelligent vision correction and management system

Abstract

The invention discloses an intelligent vision correction and management system, and relates to the technical field of vision training. The system comprises a vision detection unit, a rechecking unit, a correction training unit and a management unit, wherein the vision detection unit evaluates vision through a dynamic visual chart and gesture recognition; and the re-checking unit is combined with historical data to confirm the current vision level by an expert. The correction training unit adjusts and determines a new acousto-optic television force training scheme according to a confirmation result of an expert, and VR simulation, eye massage, ciliary muscle training and nursing light system assistance are combined. And the management unit is bound with a user account, visually displays vision changes and training schemes, and facilitates tracking and management. According to the invention, the vision detection efficiency can be improved, and the automation degree of vision training can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of visual training, and particularly to an intelligent vision correction and management system. Background Art

[0002] In the existing field of vision correction and management, traditional vision detection and correction methods have many deficiencies. The traditional vision chart detection method is single, lacking dynamics and interactivity, resulting in inaccurate detection results, and the detection process is boring, with poor user experience. In addition, correction training often relies on fixed training plans and equipment, lacking personalization and pertinence, and it is difficult to meet the actual needs of different users.

[0003] Although some vision correction systems have begun to introduce new technologies in recent years, such as electronic vision charts and intelligent training equipment, these systems still have some obvious defects. For example, some systems lack the ability to collect and recognize user actions in real time, resulting in limited detection efficiency and accuracy; some other systems lack an effective review mechanism, making the detection results may be affected by human factors. In addition, existing correction training programs are often too single, lacking diverse training modes and auxiliary equipment, and it is difficult to meet users' needs for personalized, efficient and comprehensive vision correction. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent vision correction and management system, which can improve the efficiency of vision detection and contribute to improving the automation degree of visual training.

[0005] To achieve the above purpose, the present invention provides the following solutions:

[0006] An intelligent vision correction and management system, comprising:

[0007] A vision detection unit, configured to:

[0008] Play a dynamic vision chart and collect the gesture actions of the target person in real time; the dynamic vision chart includes a vision chart image and a dynamic selection box; the dynamic selection box randomly moves to select any visual target in the vision chart image; the sizes and directions of the visual targets are arranged in a set form;

[0009] Identify the gesture actions, and after each detection cycle is completed, judge the vision according to the correct rate of each gesture to determine the vision level of the target person;

[0010] A review unit, configured to:

[0011] Extract the vision level of historical detections and send the vision level of historical detections and the currently determined vision level to the expert terminal together for confirmation by the expert terminal;

[0012] Corrective training unit for:

[0013] After receiving the confirmation signal from the expert, the previous training cycle is adjusted according to the currently determined vision level, and an updated acoustic, optical and electric vision training program is determined; wherein the acoustic, optical and electric vision training program includes: using VR to simulate real audio and video and real scenes, conducting simulated outdoor vision training, and using eye massage equipment and a light-feeding system for auxiliary processing;

[0014] Snap-in for:

[0015] The currently determined vision level and the periodically updated acoustic, optical and electronic vision training program are bound to the current login account, and all detected vision levels and the acoustic, optical and electronic vision training programs within the corresponding detection period are visualized according to a time series.

[0016] Optionally, the vision detection unit specifically includes:

[0017] A playback module, used for playing a dynamic vision chart;

[0018] Image acquisition module, used to capture the target person's gesture video in real time;

[0019] The gesture recognition module is used to input the gesture action video into the gesture recognition model for gesture recognition, and compare all gestures in a detection cycle with the selected visual targets in the dynamic vision chart of this round to obtain the gesture accuracy and the corresponding vision level of the target person.

[0020] Optionally, the vision detection unit further includes: a model training module;

[0021] The model training module is used to:

[0022] Acquire training data; the training data includes gesture videos and corresponding recognition results;

[0023] Constructing a pre-trained network; the pre-trained network includes a connected long short-term memory network and a classifier;

[0024] The training data is input into the pre-trained network, and training is performed according to a gradient descent strategy with the goal of minimizing the loss between the network output and the recognition result corresponding to the input, and the trained network is determined as the gesture recognition model.

[0025] Optionally, the review unit specifically includes:

[0026] An extraction module, used to extract the vision levels of historical tests;

[0027] The expert end confirmation module is used to send the historically detected vision level together with the currently determined vision level to the expert end and output a confirmation signal.

[0028] Optionally, the expert terminal includes at least one of a personal computer, a laptop, a smart phone, a tablet computer and a portable wearable device.

[0029] Optionally, the correction training unit specifically includes:

[0030] A signal receiving module, used for receiving a confirmation signal from the expert terminal;

[0031] A program adjustment module is used to adjust the previous training cycle based on the currently determined vision level and a set adjustment threshold to determine an updated acoustic, optical and electrical vision training program;

[0032] The program execution module is used to: perform vision correction training according to the periodically updated sound, light and electricity vision training program.

[0033] Optionally, the solution execution module specifically includes:

[0034] The eye movement submodule is used to perform eye movements based on VR videos, and train the eyeball positions of the 0° superior rectus, 45° right superior oblique, 90° right lateral rectus, 180° right inferior rectus, 125° left inferior oblique, 270° left lateral rectus, and 315° left superior oblique for 3-5 minutes.

[0035] The linear training submodule is used to conduct linear training based on VR videos. By moving a target linearly between different viewing distances, the eyeballs can undergo internal and external rotation movements within 5-7 minutes.

[0036] Yellow card training submodule, used for yellow card training based on VR video;

[0037] The auxiliary training submodule is used for auxiliary processing using eye massage equipment and light feeding system.

[0038] Optionally, the management unit specifically includes:

[0039] an association module, configured to bind the currently determined vision level and the periodically updated audio-visual vision training program to the currently logged-in account;

[0040] The visualization display module is used to visualize the vision levels of all tests and the sound, light and electricity vision training programs within the corresponding test period according to the time series.

[0041] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0042] The present invention discloses an intelligent vision correction and management system, which includes a vision detection unit, a review unit, a correction training unit and a management unit; wherein the vision detection unit evaluates vision through a dynamic vision chart and gesture recognition; the review unit combines historical data with experts to confirm the current vision level. The correction training unit adjusts and determines a new sound, light and electricity vision training program based on the expert confirmation results, combining VR simulation, eye massage, ciliary muscle training and light feeding system assistance. The management unit binds the user account to visualize vision changes and training programs for easy tracking and management. The present invention can improve the efficiency of vision detection and help improve the degree of automation of visual training. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a structural diagram of the intelligent vision correction and management system of the present invention. DETAILED DESCRIPTION

[0045] 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.

[0046] The purpose of the present invention is to provide an intelligent vision correction and management system that can improve the efficiency of vision detection and help increase the degree of automation of vision training.

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] like Figure 1 As shown, the present invention provides an intelligent vision correction and management system, including: a vision detection unit, a review unit, a correction training unit and a management unit.

[0049] A vision detection unit, configured to: play a dynamic vision chart and collect the gesture actions of the target person in real time; the dynamic vision chart includes a vision chart image and a dynamic selection box; the dynamic selection box randomly moves to select any visual target in the vision chart image; the sizes and directions of the visual targets are arranged in a set form; identify the gesture actions, and after each detection cycle is completed, determine the vision level of the target person based on the accuracy rate of each gesture.

[0050] A review unit, configured to: extract the vision level of the historical detection and send the vision level of the historical detection and the currently determined vision level to the expert terminal together for confirmation by the expert terminal.

[0051] A correction and training unit, configured to: after receiving the confirmation signal from the expert terminal, adjust the previous training cycle according to the currently determined vision level to determine the updated audio-visual-optical vision training plan for the cycle; wherein, the audio-visual-optical vision training plan includes: respectively simulating real audio-visual and real scenes through VR to perform simulated outdoor distant view training, and using an eye massage device and a low-level laser therapy system for auxiliary processing.

[0052] A management unit, configured to: bind the currently determined vision level and the updated audio-visual-optical vision training plan for the cycle to the currently logged-in account, and visually display all the detected vision levels and the corresponding audio-visual-optical vision training plans within the detection cycle according to the time series.

[0053] As a specific implementation manner, the vision detection unit specifically includes: a playback module, an image acquisition module, and a gesture recognition module.

[0054] Among them, the playback module is used to play the dynamic vision chart; the image acquisition module is used to capture the gesture action video of the target person in real time; the gesture recognition module is used to input the gesture action video into the gesture recognition model for gesture recognition, and compare all the gestures within one detection cycle with the selected visual targets in the dynamic vision chart of this round to obtain the gesture accuracy rate and the corresponding vision level of the target person.

[0055] In addition, the vision detection unit further includes: a model training module; the model training module is used to: obtain training data; the training data includes gesture videos and corresponding recognition results; construct a pre-trained network; the pre-trained network includes a connected long short-term memory network and a classifier; input the training data into the pre-trained network, aim at minimizing the loss between the network output and the input corresponding recognition result, perform training according to the gradient descent strategy, and determine the trained network as the gesture recognition model.

[0056] As a specific implementation manner, the review unit specifically includes: an extraction module and an expert terminal confirmation module.

[0057] The extraction module is used to extract the vision levels of historical tests, and the expert-side confirmation module is used to send the vision levels of historical tests together with the currently determined vision level to the expert-side and output a confirmation signal. The expert-side includes at least one of a personal computer, a laptop, a smartphone, a tablet computer, and a portable wearable device.

[0058] As a specific implementation, the correction training unit specifically includes: a signal receiving module, a plan adjustment module and a plan execution module.

[0059] Among them, the signal receiving module is used to receive the confirmation signal from the expert end; the program adjustment module is used to adjust the previous training cycle based on the currently determined vision level and the set adjustment threshold, and determine the sound, light and electricity vision training program after the cycle is updated; the program execution module is used to: perform vision correction training according to the sound, light and electricity vision training program after the cycle is updated.

[0060] The program execution module specifically includes: eye movement sub-module, straight line practice sub-module, yellow card training sub-module and auxiliary training sub-module.

[0061] The eye movement submodule is used to perform eye movements based on VR videos, training the positions of the 0° superior rectus, 45° right superior oblique, 90° right lateral rectus, 180° right inferior rectus, 125° left inferior oblique, 270° left lateral rectus, and 315° left superior oblique for 3-5 minutes. The straight line exercise submodule is used to perform straight line exercises based on VR videos, using the linear movement of targets between different viewing distances to induce inward and outward rotation of the eye within 5-7 minutes. The yellow card training submodule is used to perform yellow card training based on VR videos. The auxiliary training submodule is used to perform auxiliary processing using eye massage equipment and a light feeding system.

[0062] As a specific implementation method, the management unit specifically includes: an association module and a visualization display module.

[0063] The association module is used to bind the currently determined vision level and the periodically updated acoustic, optical and electrical vision training program to the currently logged-in account. The visualization display module is used to visualize all detected vision levels and the acoustic, optical and electrical vision training programs within the corresponding detection period according to a time series.

[0064] Based on the above technical solution, the following embodiments are provided.

[0065] This embodiment provides a specific application of an intelligent vision correction and management system. The system consists of four modules: a vision detection unit, a review unit, a correction training unit, and a management unit. These modules work together to provide users with personalized vision correction services.

[0066] In the vision detection unit, the playback module first shows a dynamic vision chart to the target person. The chart contains a vision chart image and a randomly moving dynamic selection box for selecting any visual target in the image. The image acquisition module captures the target person's gesture actions in real time and generates a gesture action video. The gesture recognition module then uses a pre-trained gesture recognition model to recognize the gestures in the video and compares them with the selected visual targets in the dynamic vision chart to calculate the gesture accuracy rate, thereby evaluating the vision level of the target person.

[0067] The model training module is responsible for optimizing the gesture recognition model. By obtaining training data containing gesture videos and corresponding recognition results, a pre-trained network including a long short-term memory network and a classifier is constructed. Subsequently, the training data is input into the network, and with the goal of minimizing the loss between the network output and the corresponding recognition result of the input, a gradient descent strategy is used for training until the model performance meets the requirements.

[0068] In the review unit, the extraction module extracts the historical vision detection data of the target person from the historical database. The expert terminal confirmation module sends this data and the currently detected vision level to the expert terminal (such as a personal computer, smartphone, etc.) for the expert to review and confirm.

[0069] After receiving the expert confirmation signal, the correction training unit is received by the signal receiving module. The plan adjustment module adjusts the previous training cycle according to the current vision level and a preset adjustment threshold to generate a new audio-visual-optical vision training plan. The plan includes multiple sub-modules such as eye movement, straight line practice, yellow card training, and auxiliary training. Through VR videos to simulate real scenarios, combined with an eye massage device and a light nourishing system for auxiliary training.

[0070] The plan execution module then guides the target person to perform vision correction training according to the adjusted training plan. For example, the eye movement sub-module will guide the eyes to perform training at specific positions based on the VR video; the straight line practice sub-module promotes the internal and external rotation movements of the eyes through the straight line movement of the target; the yellow card training sub-module uses the yellow card for visual training; the auxiliary training sub-module uses the eye massage device and the light nourishing system for additional auxiliary processing.

[0071] Among them, in the eye movement sub-module, for eye movement, strength exercises for the superior rectus muscle at 0°, the superior oblique muscle at 45°, the lateral rectus muscle at 90°, the inferior rectus muscle at 180°, the inferior oblique muscle at 125°, the lateral rectus muscle at 270°, and the superior oblique muscle at 315° of the eye are carried out, and the training time is 3 - 5 minutes to improve the fine regulation function of the eye muscles.

[0072] In the straight-line practice sub-module, through the straight-line movement of the target between different viewing distance points, slow, fast movement and stay, the eyeballs perform internal rotation movement and external rotation movement within 5 - 7 minutes, achieving the effect of strengthening the eye muscles and improving the accommodation ability.

[0073] In the yellow card training sub-module, the ultimate goal of yellow card training is to be able to see the mental image one hopes to see, that is, to stimulate internal vision. This goal is called the activation of the right brain, and through this activation, it helps to train the brain's ability for photographic reading.

[0074] The yellow card consists of a blue dot in the center and the surrounding orange-yellow color. The cone cells in our eyes are most sensitive to the four colors of yellow, blue, red, and green. Therefore, the blue and yellow colors of the yellow card, as well as the four colors of the mandala card, can directly stimulate the cone cells, allowing us to easily see complementary colors (yellow and blue are complementary colors, and red and green are also complementary colors) after closing our eyes. Yellow card training is a guiding method. Through this excellent method, it guides our high concentration of attention to the brain screen. Let us through this serious feeling and experience in the a-wave state, make this internal vision function that has sunk deep into the subconscious slowly float to our conscious mind again, so that in the waking state, it helps to achieve the purpose of high-speed reading and high-speed memory.

[0075] In the management unit, the association module binds the current vision level and training plan to the currently logged-in account to ensure the accuracy and traceability of the data. The visualization display module then visualizes all detected vision levels and the audio-visual-kinesthetic vision training plan within the corresponding detection period according to the time series, facilitating users to view and understand the progress and effect of vision correction at any time.

[0076] Therefore, the system in this embodiment plays a dynamic vision chart through the vision detection unit and real-time collects the gesture actions of the target person, thereby achieving a quick and accurate judgment of the vision level. At the same time, the review unit can extract historical detection data for experts to confirm to ensure the accuracy of the detection results. The correction training unit formulates a personalized audio-visual-kinesthetic vision training plan according to the current vision level and improves vision through VR simulation and auxiliary devices. The management unit is responsible for binding the vision level and training plan to the account, realizing the visualization display and tracking of the data. Such a system not only improves the efficiency of vision detection and correction but also provides users with a more convenient and personalized vision management service.

[0077] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other.

[0078] In this text, specific examples are used to elaborate on the principles and implementation modes of the present invention. The description of the above embodiments is only used to help understand the core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation modes and application scopes. In summary, the content of this specification should not be construed as a limitation on the present invention.

Claims

1. A vision intelligent correction and management system, characterized in that, Including: A vision detection unit, configured to: Play a dynamic visual acuity chart and collect the gesture actions of the target person in real time; the dynamic visual acuity chart includes a visual acuity chart image and a dynamic selection box; the dynamic selection box randomly moves to select any visual target in the visual acuity chart image; the sizes and directions of the visual targets are arranged in a set form; Identify the gesture actions, and after each detection cycle is completed, judge the visual acuity according to the correct rate of each gesture to determine the visual acuity level of the target person; A review unit, configured to: Extract the visual acuity level of historical detections and send the visual acuity level of the historical detections and the currently determined visual acuity level to the expert terminal together for confirmation by the expert terminal; A correction and training unit, configured to: After receiving the confirmation signal from the expert terminal, adjust the previous training cycle according to the currently determined visual acuity level to determine the updated audio-visual and optometric training plan for the cycle; wherein, the audio-visual and optometric training plan includes: respectively simulating real audio-visual and real scenes through VR to perform simulated outdoor distant vision training, and using an eye massage device and a laser system for auxiliary processing; A management unit, configured to: Bind the currently determined visual acuity level and the updated audio-visual and optometric training plan for the cycle to the currently logged-in account, and visually display all detected visual acuity levels and the corresponding audio-visual and optometric training plans within the detection cycle according to the time sequence.

2. The vision intelligent correction and management system according to claim 1, wherein The vision detection unit specifically includes: A playback module, configured to play a dynamic visual acuity chart; An image acquisition module, configured to capture a video of the gesture actions of the target person in real time; A gesture recognition module, configured to input the gesture action video into a gesture recognition model for gesture recognition, and compare all gestures within one detection cycle with the selected visual targets in the current round of the dynamic visual acuity chart to obtain the gesture correct rate and the corresponding visual acuity level of the target person.

3. The vision intelligent correction and management system according to claim 2, characterized in that The vision detection unit further includes: a model training module; The model training module is configured to: Obtain training data; the training data includes gesture videos and corresponding recognition results; Construct a pre-trained network; the pre-trained network includes a connected long short-term memory network and a classifier; Input the training data into the pre-trained network, target to minimize the loss between the network output and the input corresponding recognition result, perform training according to the gradient descent strategy, and determine the trained network as the gesture recognition model.

4. The vision intelligent correction and management system according to claim 1, characterized in that The review unit specifically includes: An extraction module, configured to extract the visual acuity level of historical detections; An expert terminal confirmation module, configured to send the visual acuity level of the historical detections and the currently determined visual acuity level to the expert terminal together and output a confirmation signal.

5. The vision intelligent correction and management system according to claim 4, characterized in that, The expert terminal includes at least one of a personal computer, a laptop computer, a smart phone, a tablet computer, and a portable wearable device.

6. The vision intelligent correction and management system according to claim 1, characterized in that The correction and training unit specifically includes: A signal receiving module, configured to receive the confirmation signal from the expert terminal; A plan adjustment module, configured to adjust the previous training cycle based on the currently determined visual acuity level according to a set adjustment threshold to determine the updated audio-visual and optometric training plan for the cycle; A program execution module, configured to: perform vision correction training according to the updated acousto-optic vision training program in a cycle.

7. The vision intelligent correction and management system according to claim 6, characterized in that, The program execution module specifically includes: An eye movement sub-module, configured to perform eye movement based on a VR video, and perform 3-5 minutes of training on the positions of the superior rectus muscle of the eye at 0°, the superior oblique muscle of the right eye at 45°, the lateral rectus muscle of the right eye at 90°, the inferior rectus muscle of the right eye at 180°, the inferior oblique muscle of the left eye at 125°, the lateral rectus muscle of the left eye at 270°, and the superior oblique muscle of the left eye at 315° respectively; A straight line practice sub-module, configured to perform straight line practice based on a VR video, and cause the eye to perform internal rotation movement and external rotation movement within 5-7 minutes through the straight line movement of a target between different viewing distance points; A yellow card training sub-module, configured to perform yellow card training based on a VR video; An auxiliary training sub-module, configured to perform auxiliary processing using an eye massage device and a red light therapy system.

8. The vision intelligent correction and management system according to claim 1, wherein The management unit specifically includes: An association module, configured to bind the currently determined vision level and the updated acousto-optic vision training program in a cycle to the currently logged-in account; A visualization display module, configured to visually display all detected vision levels and the acousto-optic vision training programs within the corresponding detection cycles according to a time series.

Images