Intelligent algorithm for correcting vision by automatically generating vision training scheme based on vision report
By generating vision reports and AI intelligent algorithms, automatically generating visual training solutions, combined with AI eye tracking and 3D video, the problems of different eye correction effects and high cost in the existing technology are solved, and efficient and personalized vision correction effects are achieved.
Patent Information
- Application Number
- CN202410845616.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing eye correction technology relies on manual labor, resulting in large differences in correction effects, high cost and low efficiency, and the visual training content is single and cannot be effectively replicated, resulting in poor visual correction effects.
Vision reports are generated through eye detection devices, and targeted visual training solutions are automatically generated using AI intelligent algorithms. Combined with AI eye tracking technology and 3D video playback, the binocular stage-free visual segmentation function and real-time picture adjustment are realized, and personalized visual training content is provided.
It improves the efficiency and effectiveness of the correction plan, reduces costs, enhances the pertinence and intensity of visual training, and ensures the user's vision correction effect.
Smart Images

Figure CN118924231A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of eye vision correction, and in particular to an intelligent algorithm for automatically generating a visual training program for correcting vision based on a vision report. Background Art
[0003] Visual training is a way of training the eyes and brain. It retrains the relationship between the brain and the eyes, just like doing eye movements. It is a continuous stimulation and training of the brain's visual neurocognitive system. The correction is designed for people with insufficient eye movement and poor eye concentration. In addition, the corrective vision training program can train and correct eye health problems such as myopia, hyperopia and amblyopia. Through visual training, it increases the eye movement, focusing, fixation ability, and binocular cooperation ability, improves visual processing ability and corrects visual functions such as amblyopia.
[0004] Nowadays, eye correction generally relies too much on corrective personnel. There are great differences in technical methods, resulting in differences in eye correction effects. Moreover, since the eye problems of users being corrected are different, the correction plan cannot be replicated, resulting in high cost and low efficiency of the correction plan. At the same time, the current visual training content is relatively simple, resulting in poor visual training effects, which leads to poor results in eye vision correction. Summary of the invention
[0005] The purpose of the present invention is to provide an intelligent algorithm for automatically generating a visual training plan to correct vision based on a vision report, so as to solve the problems raised in the above-mentioned background technology.
[0006] To achieve the above object, the present invention provides the following technical solution: an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, comprising the following steps:
[0007] S1. The patient uses an eye detection device to check the health of his eyes, and a vision report of the user is generated after the check is completed;
[0008] S2. When the user's vision report is generated, the data in the vision report is synchronously transmitted to the computer, and the computer generates a visual training plan for the corresponding user's vision report through AI intelligence;
[0009] S3. Send the generated visual training program to the visual training device, and customize the corresponding visual training correction program for the corresponding user according to his / her eye condition;
[0010] S4. The user uses the visual training device, and then the visual training device displays and plays the visual training content. During the user's training process, the visual training device uses AI intelligent eye tracking technology to adjust the image in real time to adapt to the range seen by the user's eyes;
[0011] S5. The visual training equipment sets the visual training picture according to the left and right eyes with binocular non-stepped vision function, and has the function of independently adjusting binocular picture parameters to achieve adaptive training for the left and right eyes, and the visual training equipment has a 3D video playback function;
[0012] S6. The visual training device plays a path video, a dot is set on the video screen, the dot moves according to the path rules, and the user's eyeball moves along the dot.
[0013] In a further embodiment, the visual training device has built-in training videos, and the visual training device intelligently matches the corresponding visual training videos according to the eye problems detected by the user, and the visual training device intelligently generates corresponding visual training plans according to the content of the user's vision report, thereby improving the generation of visual training plans to be more targeted and more efficient.
[0014] In a further embodiment, the eye tracking algorithm in S4 is as follows:
[0015] C = (x0, y0) - (x, y)
[0016] x0=r*sin(α)
[0017] y0=r*cos(α)
[0018] Among them, C is the moving distance, x is the starting point of the horizontal coordinate, y is the starting point of the vertical coordinate, x0 is the end point of the horizontal coordinate, y0 is the end point of the vertical coordinate, cos(α) and sin(α) are the angles of eyeball movement, and r is the radius of the eyeball.
[0019] In a further embodiment, there are two types of visual training devices, one of which is a head-mounted visual training device, and the other is a portable visual training device. The portable visual training device is a flat display device, and the display screen of the portable visual training device has a 3D display function, specifically adopting a naked-eye 3D display solution.
[0020] In a further embodiment, the portable visual training device is provided with a distance sensor. When the visual training device detects that the distance between the user's eyes and the screen is less than 40 cm, the visual training device emits an alarm to prevent the user's eyes from being too close to the screen and causing eye burden, thereby ensuring the effect of visual training in correcting vision.
[0021] In a further embodiment, the main training contents of the portable vision training equipment include training programs of eye stimulation, eye accommodation training, simultaneous vision training, fusion vision training and stereoscopic vision training. The portable vision training equipment is convenient for users to use in daily life and improves the convenience of visual training and vision correction.
[0022] In a further embodiment, the head-mounted visual training device is connected to a control computer. The main training contents of the head-mounted visual training device include four training programs: myopia prevention and control training, amblyopia training, visual fatigue training and strabismus training. The head-mounted visual training device can effectively improve the visual training effect.
[0023] In a further embodiment, the vision report includes eye degree, eye astigmatism, eye pupil distance, eye strabismus and eye amblyopia data, and the visual training device generates a corresponding visual training plan to correct vision based on the vision report data. Through various vision report data, the system can conveniently formulate visual training plans for different users.
[0024] In a further embodiment, the visual training device has built-in AI intelligent management. During the visual training process, the visual training device automatically adjusts the visual training content and visual training plan based on the content of the vision examination report of the user's positioned eyes. By regularly adjusting the visual training plan, the effect of the user's visual training plan is improved.
[0025] In a further embodiment, the binocular picture parameters adjusted in S5 include picture brightness, picture color temperature, picture contrast, picture saturation, picture sharpness blur and picture clarity. The picture parameters are adjusted in real time and intelligently according to the problems in the user's vision report. By adjusting the various parameters of the picture, some parameters that affect the user's eyes and the damage to the eyes can be reduced, ensuring that the user's vision can be further corrected in a safe visual training environment;
[0026] The specific parameter adjustment algorithm formula is as follows:
[0027] x=(α÷β)×100%
[0028] Among them, x is the parameter value percentage adjusted, α is the corresponding parameter value in the vision report, and β is the highest value of the corresponding picture parameter.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] 1. The present invention uses the vision report generated by the user's eye examination as a basis, and then customizes matching visual training plans for different users according to their actual vision conditions through AI intelligent matching. As the user's vision correction condition after visual training, the system can regularly adjust the visual training plan according to the real-time eye state, thereby reducing labor costs and making the user's visual training plan more suitable for the state of his eyes, thereby improving the efficiency and correction effect of the correction plan, while reducing the cost of formulating the correction plan, so that the visual training plan for vision correction is more suitable for the user and the visual training effect is better;
[0031] 2. The present invention uses AI intelligent visual training, 3D visual training and AI eye tracking to provide users with good visual training content, and uses real-time eye tracking to keep the 3D image in the best state, so that users can use the visual training equipment more effectively, improve the visual training intensity, ensure the effect of user visual training, and improve the effect of correcting the user's eyesight;
[0032] 3. In the present invention, the visual training device sets two sets of images for the left and right eyes and utilizes binocular infinite separation technology to make the image parameters seen by the left and right eyes different. According to the different problems and situations of the left and right eyes of the user, the image parameters seen are adjusted in a targeted manner, so that the visual device can perform targeted visual training and correct the user's vision according to the different situations and problems of the left and right eyes of the user. The targeted nature of the visual training device during the user's visual training can effectively improve the effect of the user's vision correction. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 It is a schematic diagram of the process framework of the visual training vision correction algorithm of the present invention. DETAILED DESCRIPTION
[0034] The following will be described clearly and completely in conjunction with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0035] Example 1
[0036] See also Figure 1 This embodiment provides an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, comprising the following steps:
[0037] S1. The patient uses an eye detection device to check the health of his eyes, and a vision report of the user is generated after the check is completed;
[0038] S2. When the user's vision report is generated, the data in the vision report is synchronously transmitted to the computer, and the computer generates a visual training plan for the corresponding user's vision report through AI intelligence;
[0039] S3. Send the generated visual training plan to the visual training device, customize the corresponding visual training correction plan for the corresponding user according to the eye condition of the user, and the visual training device automatically customizes the visual training plan, and then the visual training device displays the visual training plan in the form of video. The user can improve the user's eyesight during the video training process, thereby efficiently correcting the user's eyesight and improving the user's eye health and vision;
[0040] S4. The user uses the visual training device, and then the visual training device displays and plays the visual training content. During the user's practical training, the visual training device uses AI intelligent eye tracking technology to adjust the image in real time to fit the range seen by the user's eyes. In order to ensure the user's comfort during the visual training process and to ensure that the 3D visual training content does not cause strabismus to the user's eyes, the visual training device uses AI intelligent and automatic eye tracking to identify the position where the pupil is locked. The image will be adjusted in real time by AI to face the user's eyes, so that the visual training image can always be in front of the user's eyes, preventing the user from having strabismus caused by long-term tilted viewing of the eyeball due to the tilted video, and ensuring that the visual training vision correction is safer and more stable;
[0041] S5. The visual training equipment sets the visual training picture according to the left and right eyes with binocular stepless separation function, and has the function of independently adjusting binocular picture parameters to achieve adaptive training of the left and right eyes. The visual training equipment has a 3D video playback function, and the visual training picture parameters seen by the left and right eyes are independently adjusted. The binocular stepless separation solution can independently adjust the binocular picture state, making the visual training program more flexible;
[0042] S6. The visual training device plays a path video. A dot is set on the video screen. The dot moves according to the path pattern. The user's eyeball follows the dot to move. The visual training content sets a dot activity video with a regular path. The user's eyes need to follow the dot to move. By moving the eyeball, the eyeball rotates. The eyeball rotation can drive the muscle activity around the eye, so that the user can relax the eye muscles, improve eye fatigue, and play a certain role in vision correction.
[0043] The visual training device has built-in training videos. The visual training device intelligently matches the corresponding visual training video according to the eye problems detected by the user, and the visual training device intelligently generates the corresponding visual training plan according to the content of the user's vision report. The visual training device for vision correction contains at least 130 visual training videos, and these videos have different visual training correction effects, which can provide the user's eyes with myopia, hyperopia, vision adjustment, stereoscopic vision, simultaneous vision and fusion vision training, making the training content of the visual training device more comprehensive and improving the user's visual training and vision correction more comprehensive and efficient.
[0044] Example 2
[0045] See also Figure 1 , further improvements are made on the basis of Example 1.
[0046] The eye tracking algorithm in S4 is as follows:
[0047] C = (x0, y0) - (x, y)
[0048] x0=r*sin(α)
[0049] y0=r*cos(α)
[0050] Among them, C is the moving distance, x is the starting point of the horizontal coordinate, y is the starting point of the vertical coordinate, x0 is the end point of the horizontal coordinate, y0 is the end point of the vertical coordinate, cos(α) and sin(α) are the eye movement angles, r is the radius of the eyeball, first determine the eyeball position (x0, y0), and then calculate the angle a, and then calculate the position of the eyeball. Suppose the left eye is bal1 and the right eye is bal2. The center coordinates of the left eye are (39,72), and the center coordinates of the right eye are (106,68). Then detect the horizontal and vertical coordinates of the left and right eyes again, confirm the displacement number with the above center horizontal and vertical coordinates, confirm the position of the eyeball, and then use the visual training video to realize eye tracking image correction.
[0051] There are two types of visual training equipment, one is a head-mounted visual training equipment, and the other is a portable visual training equipment. The portable visual training equipment is a flat display device. The display screen of the portable visual training equipment has a 3D display function, and specifically adopts a naked-eye 3D display solution. The portable visual training equipment uses naked-eye 3D technology to enable the visual training content to display the 3D visual training effect used for visual training like a head-mounted visual training equipment, thereby enabling the portable visual training equipment to efficiently perform visual training correction on the user's eyes.
[0052] The portable visual training device is equipped with a distance sensor. When the visual training device detects that the distance between the user's eyes and the screen is less than 40 cm, the visual training device will issue an alarm to prevent the user from getting too close to the screen of the visual training device when using the visual training device, which may cause vision loss. This ensures the effect of visual training in correcting vision, reduces further damage to the eyes, and ensures that the eyes can perform vision correction stably.
[0053] The main training contents of portable vision training equipment include eye stimulation, eye accommodation training, simultaneous vision training, fusion vision training and stereoscopic vision training. Compared with head-mounted vision training equipment, portable vision training equipment is suitable for users' home vision training. It is easy to carry and has low cost. Users can purchase the product and use it for daily vision training and vision correction outside the hospital.
[0054] The head-mounted visual training device is connected to a control computer. The main training contents of the head-mounted visual training device include four training programs: myopia prevention and control training, amblyopia training, visual fatigue training and strabismus training. The head-mounted visual training device is suitable for users to perform visual training and correct vision in a correctional hospital. The head-mounted visual training device has a good visual training effect, but because of its cost and the computer equipment required for control, it is only suitable for use in visual training and correction in hospitals.
[0055] Example 3
[0056] See also Figure 1 , further improvements are made on the basis of Example 1.
[0057] The vision report includes data on eye degree, eye astigmatism, eye pupil distance, eye strabismus and eye amblyopia. Based on the above vision report data, the visual training equipment generates a corresponding visual training plan to correct the eyes. The user checks the vision status of the eyes and, based on the vision report generated by the examination, the visual training equipment system generates a visual training correction plan corresponding to the user's vision report, thereby reducing repetitive visual training plans and making the formulation of visual training plans more efficient. The visual training content is more targeted at the user's actual vision condition. Targeted vision correction can make the visual training effect more effective, and there is no need for staff participation in the process. The visual training plan is automatically generated, which can reduce manual workload and reduce the cost of formulating visual training plans.
[0058] The visual training equipment has built-in AI intelligent management. During the visual training process, the visual training equipment automatically adjusts the visual training content and plan based on the content of the vision examination report of the user's eye. As the visual training correction progresses, the user's eye vision and damage are gradually corrected. Therefore, it is necessary to adjust the visual training content to suit users at different stages. Therefore, the visual training equipment uses AI intelligent management and adjustment to automatically adjust the visual training plan appropriately during the user's visual training process, improve the visual training content of different users to better suit the user's own situation, improve the correction effect of the visual training content, and make the user's eye visual training correction more comprehensive and efficient.
[0059] The binocular picture parameters adjusted in S5 include picture brightness, picture color temperature, picture contrast, picture saturation, picture sharpness blur and picture clarity. The picture parameters are adjusted in real time and intelligently according to the problems in the user's vision report. Because the eye conditions of different users are different, and the left and right eyes of the same user also have vision differences and damage differences, when the user needs to correct vision through visual training, in order to meet the difference problem of the left and right eyes, the picture of the visual training content is set to two sets of visual training content with different parameter specifications, which can be more targeted for the eyes. Instead of correcting the vision of both eyes through the same set of visual training content, such targeted visual training content can enable the left and right eyes to perform visual training correction more fully and efficiently;
[0060] The specific parameter adjustment algorithm formula is as follows:
[0061] x=(α÷β)×100%
[0062] Among them, x is the percentage of the parameter value adjusted, α is the corresponding parameter value in the vision report, and β is the highest value of the corresponding screen parameter. The parameters detected in the vision report are used to match the screen parameter values. The corresponding parameter values are calculated by adjusting their percentages to achieve automatic adjustment of the corresponding parameters of the screen to the required parameter percentages.
[0063] The visual training program uses an intelligent algorithm for correcting vision. When in use, the user first uses the detection equipment to check his or her eyes and various functions of the eyes. After the inspection, the detection device generates a vision examination report for the user. The vision report includes the degree of the user's eyes and other health indicators. The vision report is then transmitted to the control computer of the visual training device. At this time, the intelligent AI in the computer is started, and the vision report is identified and judged by the intelligent AI. Then, based on the analyzed results of the vision report, the control computer of the visual training device generates a visual training program for the corresponding user, and synchronizes the program to the visual training device. After that, the user can use the visual training device to perform vision training correction. During the user's visual training process, the visual training device will play the visual training content according to the formulated visual training program. There are two types of visual training devices, one is a head-mounted visual training device. When in use, the user brings the head-mounted visual training device to the head, and then turns on the head-mounted visual training device. Then the user sees the visual training content from the display part of the head-mounted visual training device. The visual training content is 3D video. During the visual training process, the user will use the head-mounted visual training device to play the visual training content according to the established visual training program. AI eye tracking technology ensures that the visual training screen is always centered where the user's eyes see it, making the visual training screen more in line with the user's eyes. At the same time, another portable visual training device also has the function of centering the visual training content with eye tracking. Some users may have different degrees of myopia or other problems in their left and right eyes. Therefore, both the head-mounted visual training device and the portable visual training device can divide the visual training content screen into left and right parts, and the two parts have the ability to independently adjust the screen parameters. In the visual training process, the visual training screen content on one side can be adjusted in a targeted manner to suit the different problems of the left and right eyes, so that the user's left and right eyes can obtain more targeted and stronger visual training correction effects, and the visual training device can provide stronger visual training correction for the user's eyes. The portable visual training device can monitor the distance between the device and the user's eyeballs during use. When the user's eyes are less than 40cm away from the device, the device will issue an obvious alarm to remind the user that the eyes are too close to the display screen to prevent the eyes from being damaged again due to the close distance between the eyes and the screen, thereby ensuring the safety of the user's eyes.
[0064] Although 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 the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. An intelligent algorithm that automatically generates visual training plans to correct vision based on vision reports, It is characterized in that The following steps are involved: S1. The patient uses an eye detection device to check the health of his eyes, and a vision report of the user is generated after the check is completed; S2. When the user's vision report is generated, the data in the vision report is synchronously transmitted to the computer, and the computer generates a visual training plan for the corresponding user's vision report through AI intelligence; S3. Send the generated visual training program to the visual training device, and customize the corresponding visual training correction program for the corresponding user according to his / her eye condition; S4. The user uses the visual training device, and then the visual training device displays and plays the visual training content. During the user's training process, the visual training device uses AI intelligent eye tracking technology to adjust the image in real time to adapt to the range seen by the user's eyes; S5. The visual training equipment sets the visual training picture according to the left and right eyes with binocular non-stepped vision function, and has the function of independently adjusting binocular picture parameters to achieve adaptive training for the left and right eyes, and the visual training equipment has a 3D video playback function; S6. The visual training device plays a path video, a dot is set on the video screen, the dot moves according to the path rules, and the user's eyeball moves along the dot.
2. According to claim 1, an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, Features: The visual training device has a built-in training video, and the visual training intelligently matches the corresponding visual training video according to the eye problems detected by the user, and the visual training device intelligently generates a corresponding visual training plan according to the content of the user's vision report.
3. According to claim 1, an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, It is characterized in that The eye tracking algorithm formula in S4 is as follows: C = (x0, y0) - (x, y) x0=r*sin(α) y0=r*cos(α) Among them, C is the moving distance, x is the starting point of the horizontal coordinate, y is the starting point of the vertical coordinate, x0 is the end point of the horizontal coordinate, y0 is the end point of the vertical coordinate, cos(α) and sin(α) are the angles of eyeball movement, and r is the radius of the eyeball.
4. According to claim 1, an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, Features: There are two types of visual training devices, one of which is a head-mounted visual training device, and the other is a portable visual training device. The portable visual training device is a flat display device. The display screen of the portable visual training device has a 3D display function, specifically adopting a naked-eye 3D display solution.
5. According to claim 4, an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, Features: The portable visual training device is provided with a distance sensor. When the visual training device detects that the distance between the user's eyes and the screen is less than 40 cm, the visual training device emits an alarm sound.
6. The intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report according to claim 4, Features: The main training contents of the portable vision training equipment include training programs of eye stimulation, eye accommodation training, simultaneous vision training, fusion vision training and stereoscopic vision training.
7. The intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report according to claim 4, Features: The head-mounted visual training device is connected to a control computer, and the main training contents of the head-mounted visual training device include four training programs: myopia prevention and control training, amblyopia training, visual fatigue training and strabismus training.
8. According to claim 1, an intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report, Features: The vision report includes data on eye degree, eye astigmatism, eye pupil distance, eye strabismus and eye amblyopia, and the vision training device generates a corresponding vision training plan to correct vision according to the vision report data.
9. The intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report according to claim 1, Features: The visual training device is equipped with AI intelligent management. During the visual training process, the visual training device automatically adjusts the visual training content and plan according to the content of the vision examination report of the user's positioned eyes.
10. The intelligent algorithm for automatically generating a visual training program to correct vision based on a vision report according to claim 1, Features: The binocular image parameters adjusted in S5 include image brightness, image color temperature, image contrast, image saturation, image sharpness and blur, and image clarity. The image parameters are adjusted in real time and intelligently according to the problems in the user's vision report; The specific parameter adjustment algorithm formula is as follows: x=(α÷β)×100% Among them, x is the parameter value percentage adjusted, α is the corresponding parameter value in the vision report, and β is the highest value of the corresponding picture parameter.
