Wearable visual training equipment based on artificial intelligence

Abstract

The invention relates to the technical field of auxiliary medical instruments, in particular to wearable visual training equipment based on artificial intelligence, which comprises a shell, a fixing assembly and symmetrically arranged display assemblies are arranged on the shell, adjusting assemblies are arranged on the display assemblies, and an audio output module is further arranged on the shell; the control system is used for controlling the display assembly to work, providing different test images for the eyes of the user, collecting audio information output by the user, obtaining description features according to the audio information, calculating the similarity between the description features and features in the test images, determining that the visual function of the user is normal when the similarity is larger than a preset value, and otherwise, determining that the visual function of the user is normal. When the description features are smaller than or equal to the preset numerical value, the visual function of the user is abnormal, at the moment, a difference image between the description features and the features in the test image is generated, and the visual dysfunction type of the user is judged according to the difference image, and the method and device are used for judging the visual dysfunction type of the user and formulating a corresponding visual training scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of auxiliary medical devices, and in particular to wearable vision training equipment based on artificial intelligence. Background Art

[0002] Vision training, first known as "orthopia", was proposed by French ophthalmologist Professor Javal. It is mainly used as an alternative to surgery to eliminate strabismus and other binocular vision and binocular movement abnormalities. Over time, vision training technology has continued to develop and integrate knowledge from multiple disciplines, including ophthalmology, optometry, physiology, etc. Since the 20th century, with the advancement of ophthalmic medical technology and the in-depth study of visual science, vision training has gradually become an important non-surgical treatment method for improving various vision problems.

[0003] Visual training is mainly divided into two types: home training and training room training. Home training is carried out in a home environment, and the training products are simple to operate and have a wide range of applications, such as threading beads, tracing, plug-in boards, etc. The advantage of home training is that it is convenient and flexible, and can be carried out at any time in daily life, but it requires supervision and guidance from parents or guardians. Training in the training room is carried out under the guidance of professional trainers, and the operation of training products is carried out in an interactive and experiential way. Training in the training room is more systematic and personalized, and can formulate targeted training plans based on the specific situation of the patient, and adjust the training methods and plans at different stages according to the training effect. Training in the training room usually uses tools such as two-sided mirrors, vision cards, and multimedia visual function training and treatment systems.

[0004] Artificial Intelligence (AI) refers to enabling machines to "think" and "learn" like humans, processing various information such as language, audio, images, and videos, and intelligently learning and inferring from them. It is not just a single technology, but a combination of multiple technologies and algorithms including deep learning, machine learning, computer vision, and natural language processing.

[0005] Existing vision training technology requires manual detection and judgment of patients' visual dysfunction in order to develop corresponding vision training plans. This process is difficult for patients to complete alone, and most existing vision training devices can only provide patients with a single visual training content, which is difficult to meet the different needs of patients. Summary of the invention

[0006] In order to solve the above problems, the present invention provides a wearable vision training device based on artificial intelligence, which is used to determine the type of visual dysfunction of the user and formulate a corresponding vision training plan.

[0007] In order to achieve the above-mentioned purpose, the technical scheme of the present invention is as follows: a wearable visual training device based on artificial intelligence, comprising a shell, a fixing component and a symmetrically arranged display component are provided on the shell, the fixing component is used to limit the relative position of the display component and the user, the display component is used to project an image to the user's eyeball, the display component is provided with an adjustment component, the adjustment component is used to adjust the angle between the corresponding display component and the user's eyeball, and the shell is also provided with an audio output module;

[0008] It also includes a control system, which is used to control the operation of the display component, provide different test images to the user's eyes, control the operation of the audio output module, guide the user to describe the image, and collect audio information output by the user, and analyze the audio information, obtain description features based on the audio information, and calculate the similarity between the description features and the features in the test image. When the similarity is greater than a preset value, the user's visual function is normal, and when it is less than or equal to the preset value, the user's visual function is abnormal. At this time, a difference image between the description features and the features in the test image is generated, and the type of visual dysfunction of the user is determined based on the difference image, and a visual training plan is formulated based on the type of visual dysfunction of the user, and a virtual image is generated based on the visual training plan, and the display component is controlled to work, and the virtual image is projected onto the user's eyeball for visual training. The control system is also used to generate a predicted motion trajectory based on the test image and the virtual image, and collect the user's eye movement information at the same time, generate an actual motion trajectory based on the user's eye movement information, and control the adjustment component and the audio output module to work when the actual motion trajectory cannot coincide with the predicted motion trajectory.

[0009] Furthermore, the display components all include a display screen, a lens is provided on the side of the display screen close to the patient, and the lens and the display screen are both installed on the inner side of the shell.

[0010] Furthermore, the adjustment components include a micropump and several airbags. The airbags are arranged along the circumference of the display screen and are connected to the micropump. An electromagnetic valve is provided at the connection between the airbag and the micropump. The display screen and the shell are ball-jointed. The control system controls the operation of the electromagnetic valve and the micropump according to eye movement information and virtual images.

[0011] Furthermore, a battery is provided on the shell, the lens is spherically hinged to the shell, and the hinge position of the lens and the shell corresponds to the hinge position of the display screen and the sphere, the lens is provided with a plurality of first permanent magnets, the first permanent magnets are arranged along the circumference of the lens, the shell is provided with a plurality of first electromagnets corresponding to the first permanent magnets, the airbags are provided with metal strips, and the metal strips are slidably fitted with metal rings, the metal rings are fixedly connected to the side walls of the display screen, the metal rings are electrically connected to the corresponding first electromagnets, and the metal strips are electrically connected to the battery.

[0012] Further, the control system includes a camera, a pickup, and a controller;

[0013] The camera is used to collect image information of the user's eyes;

[0014] The microphone is used to collect the voice information sent by the user;

[0015] The controller is used to control the operation of the display screen, provide different test images to the user's eyes, control the operation of the audio output module, guide the user to describe the image, obtain the description features according to the voice information, calculate the similarity between the description features and the features in the test image, when the similarity is greater than the preset value, the user's visual function is normal, when the similarity is less than or equal to the preset value, the user's visual function is abnormal, at this time, a difference image between the description features and the test image is generated, and the user's visual function type is judged according to the difference image, a visual training plan is formulated according to the user's visual impairment type, and a virtual image is generated according to the visual training plan, and the display screen is controlled to project the virtual image to the user's eyeball. The controller is also used to generate a predicted motion trajectory according to the test image and the virtual image, and at the same time, according to the image information, a coordinate system is constructed with the center of the user's eyeball as the origin to obtain the actual motion trajectory of the user's eyeball movement. When the actual motion trajectory does not correspond to the predicted motion trajectory, the micro pump, the solenoid valve and the audio output module are controlled to operate.

[0016] Furthermore, the controller is also used to control the display screen to work again after playing the virtual image in the visual training program, provide different test images to the user's eyes, construct review features based on audio information, calculate the similarity between the review features and the features in the test image, and stop the visual training when the similarity is greater than a preset value. When the similarity is less than or equal to the preset value, formulate a visual training program again based on the difference between the review image and the test image, and control the display screen to work again to project the virtual image to the user's eyeballs.

[0017] Furthermore, the control system also includes a temperature sensor, which is used to collect temperature information of a side wall of the shell close to the user, and the controller is used to control the operation of the display screen according to the temperature information.

[0018] Furthermore, the fixing component includes a fixing strip made of elastic material, which is fixedly connected to a side of the shell close to the user's eyeball. The shell is also provided with a plurality of fixing straps, which are used to fix the relative position of the shell and the user.

[0019] Furthermore, it also includes a communication module. The controller is used to control the communication module to output the type of visual dysfunction and the visual training plan to the user after formulating the visual training plan.

[0020] Furthermore, the controller is also used to adjust the visual training program according to the voice information.

[0021] Technical principles and beneficial effects of the above scheme:

[0022] 1. This solution provides different images for the user's two eyes, judges the characteristics of the image viewed by the user according to the user's verbal description, compares the characteristics described by the user with the characteristics of the test image, judges whether the user has visual dysfunction according to the similarity between the two, and formulates a corresponding visual training plan for the user according to the type of visual dysfunction after the user has visual dysfunction, and provides virtual images to the patient through a display screen according to the visual training plan to guide the patient to perform visual training. Compared with the prior art, this solution can independently judge the visual function state and formulate a suitable visual training plan, so that the user can independently complete visual function testing and training through this device, reducing labor costs. At the same time, because this solution uses a display screen to project test images and virtual images to the user, this solution is not limited by the use site. Compared with the prior art, visual training can be performed more conveniently and quickly, thereby improving user acceptance.

[0023] 2. The design of the adjustment component in this scheme collects the user's eye image, obtains the user's eye movement trajectory, and compares the user's eye movement trajectory with the virtual image movement trajectory. When the user's eye movement trajectory cannot coincide with the virtual image movement trajectory, the user is not looking at the test image or the virtual image. At this time, if a visual function test is being conducted, the test result obtained is not accurate. If visual training is being conducted, the training effect is not good. At this time, the angle of the display screen is adjusted by the adjustment component so that the test image or virtual image can be projected to the appropriate position of the user's eye. Compared with the prior art, this device can adjust the angle between the image and the user according to the user's eye movement trajectory, so that the image is always in the best position during the test or training process, thereby improving the accuracy and effectiveness of the test. When facing some children with weak self-control, it can also maintain certain detection and training functions.

[0024] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 An axonometric diagram of an embodiment of a wearable vision training device based on artificial intelligence of the present invention;

[0026] Figure 2 A schematic diagram of wearing an embodiment of a wearable vision training device based on artificial intelligence of the present invention;

[0027] Figure 3 A rear view of an embodiment of the wearable vision training device based on artificial intelligence of the present invention;

[0028] Figure 4 for Figure 3 Sectional view at AA;

[0029] Figure 5 for Figure 4 The enlarged view of point B in the middle;

[0030] Figure 6 This is a schematic diagram of the connection of components of an embodiment of a wearable vision training device based on artificial intelligence of the present invention;

[0031] Figure 7 A bottom view of the housing in Embodiment 6 of the embodiment of the wearable vision training device based on artificial intelligence of the present invention;

[0032] Figure 8 for Figure 7 Enlarged view of point C in the middle.

[0033] The figure marks in the drawings of the specification include: 1. shell; 2. fixing component; 21. fixing bar; 22. fixing belt; 3. display component; 31. lens; 32. display screen; 4. adjustment component; 41. battery; 42. metal ring; 43. metal bar; 44. airbag; 45. first electromagnet; 46. first permanent magnet; 5. window; 6. baffle; 61. pull rod; 62. second permanent magnet; 63. second electromagnet. DETAILED DESCRIPTION

[0034] The technical solution of the present invention will be described clearly and completely below 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 creative work are within the scope of protection of the present invention.

[0035] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0036] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] The following is further described in detail through specific implementation methods:

[0038] Embodiment 1:

[0039] As attached Figure 1 -Attached Figure 6 As shown: A wearable visual training device based on artificial intelligence includes a shell 1, on which a fixing component 2 and a symmetrically arranged display component 3 are provided, the fixing component 2 is used to limit the relative position of the display component 3 and the user, the fixing component 2 includes a fixing strip 21 made of elastic material, the fixing strip 21 is bonded and fixed to the side of the shell 1 close to the user's eyeball, and the shell 1 is also provided with a plurality of fixing belts 22, one end of the fixing belts 22 are bonded and fixed to the side wall of the shell 1, and adjacent fixing belts 22 are detachably connected by buckles, and the fixing belts 22 are used to fix the relative position of the shell 1 and the user.

[0040] The display assembly 3 is used to project an image to the user's eyeball. The display assembly 3 includes a display screen 32. A lens 31 is provided on the side of the display screen 32 close to the patient. The lens 31 and the display screen 32 are both installed on the inner side of the housing 1. The display assembly 3 is provided with an adjustment assembly 4. The adjustment assembly 4 is used to adjust the angle between the corresponding display assembly 3 and the user's eyeball. The adjustment assembly 4 includes a micro pump (not shown in the figure) and a plurality of air bags 44. The air bags 44 are arranged along the circumference of the display screen 32. The air bags 44 are bonded and fixed to the side of the display screen 32 away from the lens 31, and the air bags 44 are connected to the micro pump. A solenoid valve is provided at the connection point between the airbag 44 and the micropump, the display screen 32 is ball-hinged with the shell 1, and an audio output module is fixedly connected to the shell 1 by bolts; it also includes a control system, which includes a camera, a microphone and a controller; the camera and the controller are fixedly connected to the inner wall of the shell 1 by bolts, and the microphone is fixedly connected to the outer wall of the shell 1 by bolts. The camera is used to collect image information of the user's eyeballs, and the microphone is used to collect voice information emitted by the user. The camera, the microphone, the micropump, the solenoid valve, the display screen 32 and the audio output module are all electrically connected to the controller.

[0041] The controller is used to control the operation of the display screen 32, provide different test images to the user's eyes based on the display screen 32, and control the operation of the audio output module to guide the user to describe the image, and obtain the user's description features based on the voice information, and calculate the similarity between the description features and the features in the test image. When the similarity is greater than a preset value, the user's visual function is normal, and when the similarity is less than or equal to the preset value, the user's visual function is abnormal. At this time, a difference image between the description features and the test image is generated, and the user's visual function type is judged based on the difference image, a visual training plan is formulated according to the user's visual impairment type, and a virtual image is generated according to the visual training plan, and the display screen 32 is controlled to project the virtual image onto the user's eyeball. The controller is also used to generate a predicted motion trajectory based on the test image and the virtual image, and at the same time, a coordinate system is constructed with the user's eye center as the origin based on the image information to obtain the actual motion trajectory of the user's eyeball movement. When the actual motion trajectory does not correspond to the predicted motion trajectory, the micropump, the solenoid valve and the audio output module are controlled to operate.

[0042] The specific implementation process is as follows: when using the device, the user places the shell 1 at a suitable position on the eye, then adjusts the fixing belt 22 to fix the shell 1 at the corresponding position, presses the shell 1 to deform the fixing strip 21 and fit it to the user's face to prevent light from entering the shell 1, then starts the device, the controller controls the display screen 32 to work, the image of the display screen 32 is projected onto the patient's eyeball through the lens 31, the display screen 32 provides different images for the patient's two eyes, and the fused image of the two is the test image, and controls the audio output module to work, playing "Please describe the number, color, shape and position of the images you see", and uses the microphone to The voice uttered by the user is collected, that is, the voice information is collected, and the description features are generated according to the user's description of the image he sees in the voice information. At the same time, the features in the test image are extracted using a deep learning model, and the similarity between the description features and the features in the test image is calculated. When the similarity is greater than the set value, the image described by the user is the same as the test image, and the user's visual function is in a normal state. If the similarity is less than or equal to the set value, the image described by the user is different from the test image, that is, the user's binocular vision function is in an abnormal state. At this time, a difference image between the two is generated, and the type of visual dysfunction of the user is judged based on the difference image.

[0043] For example, the image provided for the left eye is red dots arranged up and down, while the image provided for the right eye is green dots arranged from left to right, and after the two are merged, the red dot at the bottom will merge with the green dot. Therefore, the test image is four dots, and the colors are, from top to bottom, one red in the first row, two green in the second row, and one white in the third row. The features extracted from them also correspond to them. If the descriptive features extracted from the user's voice information are four dots, and the colors are, from top to bottom, one red in the first row, two green in the second row, and one white in the third row, the descriptive features are highly similar to the features in the test image. At this time, the user's visual function is normal. If the descriptive features extracted from the user's voice information are "two dots, arranged up and down, both red", "three dots, arranged left and right, and bottom, all green", ", "five dots, red on the top, green on the left and right, one red and one green on the bottom", "five dots, three green on the left and two red on the right", and "five dots, two red on the left and three green on the right", the similarity between the descriptive features and the test image is low, and the user has obvious visual dysfunction. At this time, a difference image between the descriptive features and the test image is generated, and the difference images corresponding to the above descriptive features are, respectively, "missing the green dots located at the lower left and right", "missing the red dots arranged above and below", "the positions of the red dots below do not overlap with the green dots", "the red dot is located on the right side of the green dot", and "the red dot is located on the left side of the green dot". The above difference images respectively indicate that the user may have left eye suppression, right eye suppression, binocular fusion dysfunction, entropy and exotropia.

[0044] After judging the type of visual dysfunction of the user by describing the features and the test image, a suitable visual training program is matched for the user according to the type of visual dysfunction of the user, and a virtual image is generated according to the visual training program. For example, when the user has monocular suppression, the controller controls the display screen 32 on the suppressed eye side to work, and provides the eyeball on that side with an image that moves up, down, left, and right to stimulate the macula of the suppressed eye. When the user has binocular fusion dysfunction, the virtual image produced by the controller is a pencil, and the position of the pencil in the display screens 32 on both sides is adjusted to guide the user to perform visual fusion training. If the user has heterophoria, a corresponding virtual image is generated at this time, and the position of the virtual image is adjusted through the display screen 32 to guide the user to perform rehabilitation training.

[0045] During the testing and training process, the controller generates a predicted motion trajectory of the test image or virtual image based on the test image or virtual image, and the camera continuously collects image information of the user's eyeball, locates the user's pupil based on the image information, constructs a coordinate system with the user's pupil center as the origin, generates a trajectory of the user's eyeball movement, i.e., the actual motion trajectory, and compares the actual motion trajectory with the predicted motion trajectory frame by frame. When the actual motion trajectory does not correspond to the predicted motion trajectory, i.e., the actual motion trajectory deviates from the predicted motion trajectory, the user is not looking at the test image or virtual image. If the test is being conducted at this time, the descriptive features obtained from the voice information may be different from the information obtained when the user normally looks at the test pattern, i.e., the test results obtained are inaccurate. If visual training is being conducted, since the user is not looking at the virtual image at this time, the virtual image has a weak guiding effect on the user's eyeball, and it is difficult to achieve the effect of visual training.

[0046] At this time, the controller calculates the approximate position of the optical axis according to the position of the user's pupil, etc. in the image information, and controls the micropump and the corresponding solenoid valve to pump gas into the airbag 44 at the corresponding position, inflating part of the airbag 44, driving the display screen 32 to tilt, so that the center position of the display screen 32 coincides with the user's optical axis position, so that the virtual image is projected onto the user's eyeball more clearly and accurately, improving the test and training effects. At the same time, the audio output module is controlled to output corresponding audio to the user, prompting the user to process, such as concentrating or repeating the test, repeating the training, etc., to ensure the accuracy of the test and the effectiveness of the training.

[0047] Embodiment 2:

[0048] The difference from Example 1 lies in that a battery 41 is provided on the shell 1, the lens 31 is spherically hinged to the shell 1, and the hinge position of the lens 31 and the shell 1 corresponds to the hinge position of the display screen 32 and the sphere, a plurality of first permanent magnets 46 are bonded and fixed to the lens 31, and the first permanent magnets 46 are all arranged circumferentially along the lens 31, a plurality of first electromagnets 45 corresponding to the first permanent magnets 46 are fixed to the shell 1 by bolts, a metal strip 43 is bonded and fixed to the airbag 44, and a metal ring 42 is slidably fitted on the metal strip 43, and the material of the metal ring 42 is an elastic material, so that the metal ring 42 can be deformed following the movement of the display screen 32, so as to always fit with the surface of the metal strip 43, the metal ring 42 is bonded and fixed to the side wall of the display screen 32, the metal ring 42 is electrically connected to the corresponding first electromagnet 45, and the metal strip 43 is electrically connected to the battery 41.

[0049] The specific implementation process is as follows: In the prior art, it is often necessary to adjust the angle of the lens 31 according to the position of the display screen 32, so as to ensure the quality of the image generated by the display screen 32 after being refracted by the lens 31 and formed at the fundus of the user during use. In the process of using the present device, the battery 41 supplies power to the first electromagnet 45. At this time, the display screen 32 is parallel to the inner wall of the shell 1. The lengths of the metal strips 43 connected to the circuits of the first electromagnets 45 are the same, and the currents passing through the first electromagnets 45 are also the same, so the magnetic forces generated are also the same, and the effects on the corresponding first permanent magnets 46 are also the same, so that the lens 31 is also parallel to the inner wall of the shell 1. In the process of the controller controlling the operation of the micro pump and the solenoid valve, the display screen 32 tilts, and the tilted side and the side corresponding to it drive the metal ring 42 to slide on the metal strip 43, so that the resistance of the metal strips 43 on both sides connected to the circuits of the first electromagnets 45 changes, thereby the current flowing through the corresponding first electromagnets 45 The magnitude of the current also changes accordingly, and the magnetic force of the first electromagnet 45 also changes accordingly. As the magnetic force of the first electromagnet 45 changes, the force of each first electromagnet 45 on the corresponding first permanent magnet 46 changes, thereby driving the lens 31 to rotate and return to a state parallel to the display screen 32. This solution uses the design of the metal strip 43 and the like to change the magnitude of the current passing through the first electromagnet 45 through the movement of the display screen 32, thereby adjusting the force with which the first electromagnet 45 drives the first permanent magnet 46 to move. In the process of the controller adjusting the angle of the display screen 32 through the micropump and the first electromagnet 45, the lens 31 is adjusted synchronously to avoid the risk of distortion caused by the angle between the lens 31 and the display screen 32, which helps to improve the user experience. At the same time, compared with the solution of designing two sets of drive components for the display screen 32 and the lens 31, this solution requires less space, which helps to reduce the weight of the equipment and further improve the user experience. It also helps to reduce the amount of calculation of the controller and reduce the cost of the equipment.

[0050] Embodiment 3:

[0051] The difference from Example 2 is that the controller is also used to control the display screen 32 to work again after playing the virtual image in the visual training program, provide different test images to the user's eyes, construct review features based on audio information, calculate the similarity between the review features and the features in the test image, and stop the visual training when the similarity is greater than a preset value. When the similarity is less than or equal to the preset value, formulate a visual training program again according to the difference between the review image and the test image, and control the display screen 32 to work again to project the virtual image to the user's eyeballs.

[0052] The specific implementation process is as follows: after the user completes the visual training program, the controller controls the display screen 32 to work again, provides the test image for the user's eyes again, and receives the audio information sent by the user again, uses the audio information to obtain the review features, and calculates the similarity between the review features and the features in the test image. When the features described by the user are the same as the features in the test image, that is, the similarity is large, the user's binocular vision function is restored and the visual training is stopped. If the features described by the user are different from the features in the test image, that is, the similarity is small, the user's binocular vision function is still in an abnormal state, and a new visual training program is formulated according to the difference between the review image and the test image, and a virtual image is projected to the user's eyeballs through the display screen 32 according to the visual training program to help the user perform visual training.

[0053] Embodiment 4:

[0054] The difference from Example 3 is that the control system further includes a temperature sensor, which is fixedly connected to the housing 1 by bolts, and is used to collect temperature information of the side wall of the housing 1 close to the user, and the temperature sensor is electrically connected to the controller, and the controller is used to control the display screen 32 to work according to the temperature information.

[0055] The specific implementation process is as follows: During the wearing of the device, the temperature sensor continuously collects the temperature information of the side wall of the shell 1 close to the user, and a closed space is formed between the face and the device, which cannot effectively dissipate heat. After the user wears the device, the user's body temperature is transferred to the shell 1, causing the temperature of the shell 1 to rise. When the temperature information rises to the set temperature, the controller controls the display screen 32 to start working and play the test image. If the user takes off the device during use, the temperature transferred to the shell 1 by the user gradually decreases because the device is separated from the user. That is, when the temperature information decreases, the controller controls the display screen 32 to stop working, reducing the energy consumption of the device. Compared with the prior art, the opening and closing of the device does not require manual adjustment by the user, which can avoid the user failing to close the device in time after the end of use, thereby causing energy waste. At the same time, during the operation of the device, as the device runs, the accumulated heat in it gradually increases. This heat is transferred to the shell 1, causing the temperature information to rise. When the temperature information rises to the set value, the device may overheat at this time, and the controller cuts off the power supply of the display screen 32, thereby protecting the display screen 32.

[0056] Embodiment 5:

[0057] The difference from Example 4 is that it also includes a communication module, which is fixedly connected to the inner wall of the shell 1 by bolts. The controller is used to control the communication module to output the type of visual dysfunction and the visual training plan to the user after formulating the visual training plan. The controller is also used to adjust the visual training plan according to the voice information.

[0058] The specific implementation process is as follows: when using this device, after formulating a visual training plan corresponding to the user, the controller controls the communication module to output the corresponding visual dysfunction type and visual training plan to the user, and archives the record for subsequent query. The user can adjust the visual training content according to the content of the visual training plan by outputting voice information, such as "adjust three times to two times", and during the visual training process, the ongoing visual training can also be adjusted by outputting voice information, such as "speed up", "stop now", etc., so that the device can interact with the user and enhance the user experience of the device.

[0059] Embodiment 6:

[0060] As attached Figure 7 -Attached Figure 8 As shown, the difference from Example 5 is that a plurality of viewing windows 5 are provided on the side wall of the shell 11, and a pull rod 61 is slidably connected to the viewing windows 5, a plurality of baffles 6 are rotatably engaged on the viewing windows 5, one end of the baffles 6 is rotatably engaged with the pull rod 61, and the other end of the baffles 6 is rotatably engaged with the side wall of the window 5, a second permanent magnet 62 is provided on the pull rod 61, and a second electromagnet 63 is provided on the side wall of the window 5, and the second electromagnet 63 is electrically connected to the controller.

[0061] The specific implementation process is as follows: when using the device, the user first cuts off the power supply of the device. At this time, the second electromagnet 63 cannot attract the second permanent magnet 62, and the baffle 6 sags due to gravity, thereby exposing the window 5. The user can adjust the position of the device with the help of light entering the inside of the device from the window 5. After the adjustment is completed, the user can start the device. After the device is started, the controller connects to the second electromagnet 63, and the second electromagnet 63 generates magnetic force, driving the second permanent magnet 62 to move, thereby causing the pull rod 61 to move, driving the baffle 6 to rotate, reducing the angle between the baffle 6 and the side wall of the shell 1, thereby blocking the window 5 to prevent external light from interfering with the image display during subsequent use.

[0062] Compared with the prior art, this solution facilitates adjustment of the device when worn through the design of window 5, thereby helping to improve the wearing comfort of the device. It also enables users to more conveniently adjust the relative position of the display screen and the eyeball, so that the device can adapt to the needs of different users.

[0063] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A wearable vision training device based on artificial intelligence, comprising a housing (1), characterized in that: A housing (1) is provided with a fixing component (2) and a symmetrically arranged display component (3); the fixing component (2) is used to limit the relative position of the display component (3) and a user; the display component (3) is used to project an image onto the user's eyeball; the display component (3) is provided with an adjustment component (4); the adjustment component (4) is used to adjust the angle between the corresponding display component (3) and the user's eyeball; and the housing (1) is also provided with an audio output module; The system also includes a control system, which is used to control the operation of the display component (3), provide different test images to the user's eyes based on the display component (3), control the operation of the audio output module, guide the user to describe the image, collect audio information output by the user, analyze the audio information, obtain description features based on the audio information, calculate the similarity between the description features and the features in the test image, when the similarity is greater than a preset value, the user's visual function is normal, when it is less than or equal to the preset value, the user's visual function is abnormal, and then generate a difference image between the description features and the features in the test image, and judge the type of visual dysfunction of the user based on the difference image, formulate a visual training program based on the type of visual dysfunction of the user, and generate a virtual image based on the visual training program, control the operation of the display component (3), project the virtual image to the user's eyeball for visual training, and the control system is also used to generate a predicted motion trajectory based on the test image and the virtual image, and collect the user's eye movement information at the same time, generate an actual motion trajectory based on the user's eye movement information, and control the adjustment component (4) and the audio output module to operate when the actual motion trajectory cannot coincide with the predicted motion trajectory.

2. The wearable vision training device based on artificial intelligence according to claim 1, characterized in that: The display assembly (3) comprises a display screen (32), a lens (31) is provided on the side of the display screen (32) close to the patient, and the lens (31) and the display screen (32) are both mounted on the inner side of the housing (1).

3. The wearable vision training device based on artificial intelligence according to claim 2, characterized in that: The regulating components (4) each include a micro pump and a plurality of air bags (44), the air bags (44) are arranged along the circumference of the display screen (32), and the air bags (44) are connected to the micro pumps, and a solenoid valve is provided at the connection point between the air bags (44) and the micro pumps. The display screen (32) and the housing (1) are ball-jointed, and the control system controls the solenoid valve and the micro pump to work according to the eye movement information and the virtual image.

4. The wearable vision training device based on artificial intelligence according to claim 3 is characterized in that: A storage battery (41) is provided on the housing (1); the lens (31) is spherically hinged to the housing (1); and the hinged position of the lens (31) and the housing (1) corresponds to the hinged position of the display screen (32) and the spherical body; a plurality of first permanent magnets (46) are provided on the lens (31); the first permanent magnets (46) are arranged along the circumference of the lens (32); a plurality of first electromagnets (45) corresponding to the first permanent magnets (46) are provided on the housing (1); a metal strip (43) is provided on the airbag (44); and a metal ring (42) is slidably matched on the metal strip (43); the metal ring (42) is fixedly connected to the side wall of the display screen (32); the metal ring (42) is electrically connected to the corresponding first electromagnet (45); and the metal strip (43) is electrically connected to the storage battery.

5. The wearable vision training device based on artificial intelligence according to claim 4, characterized in that: The control system includes a camera, a pickup, and a controller; The camera is used to collect image information of the user's eyes; The microphone is used to collect the voice information sent by the user; The controller is used to control the operation of the display screen (32), provide different test images to both eyes of the user based on the display screen (32), control the operation of the audio output module, guide the user to describe the image, obtain the description feature based on the voice information, calculate the similarity between the description feature and the feature in the test image, when the similarity is greater than a preset value, the user's visual function is normal, when the similarity is less than or equal to the preset value, the user's visual function is abnormal, at this time, generate a difference image between the description feature and the test image, judge the user's visual function type based on the difference image, formulate a visual training program based on the user's visual impairment type, generate a virtual image based on the visual training program, control the display screen (32) to project the virtual image onto the user's eyeball, and the controller is also used to generate a predicted motion trajectory based on the test image and the virtual image, and at the same time, construct a coordinate system based on the image information with the user's eyeball center as the origin, obtain the actual motion trajectory of the user's eyeball movement, and when the actual motion trajectory does not correspond to the predicted motion trajectory, control the micro pump, the solenoid valve and the audio output module to operate.

6. The wearable vision training device based on artificial intelligence according to claim 5, characterized in that: The controller is also used to control the display screen (32) to work again after playing the virtual image in the visual training program, provide different test images to the user's eyes, construct a review feature based on the audio information, calculate the similarity between the review feature and the feature in the test image, stop the visual training when the similarity is greater than a preset value, and formulate a visual training program again based on the difference between the review image and the test image when the similarity is less than or equal to the preset value, and control the display screen (32) to work again to project the virtual image to the user's eyeballs.

7. The wearable vision training device based on artificial intelligence according to claim 6, characterized in that: The control system also includes a temperature sensor, which is used to collect temperature information of a side wall of the housing (1) close to a user, and the controller is used to control the operation of the display screen (32) according to the temperature information.

8. The wearable vision training device based on artificial intelligence according to claim 7, characterized in that: The fixing assembly (2) comprises a fixing strip (21) made of elastic material, the fixing strip (21) being fixedly connected to a side of the housing (1) close to the user's eyeball, and a plurality of fixing straps (22) are also provided on the housing (1), the fixing straps (22) being used to fix the relative position of the housing (1) and the user.

9. The wearable vision training device based on artificial intelligence according to claim 8, characterized in that: It also includes a communication module. The controller is used to control the communication module to output the type of visual dysfunction and the visual training program to the user after formulating the visual training program.

10. The wearable vision training device based on artificial intelligence according to claim 9, characterized in that: The controller is also used to adjust the visual training program according to the voice information.

Images