A head-mounted visual inspection and visual training device

Abstract

The present invention provides a head-mounted vision detection and training device, which is a head-mounted device characterized by including a left-eye visual axis measurement module; a right-eye visual axis measurement module; a suppression module; a distance measurement module; and an eye movement point analysis module. The present invention has the beneficial effect of simultaneously training monocular and binocular vision, allowing doctors to objectively assess the status and effectiveness of binocular vision training based on eye movement data. Furthermore, the training content can take the form of images, videos, animations, games, and other forms, and can be interactive through visual lines, increasing the fun of training and enhancing patient motivation, thereby facilitating better visual training results.

Description

[0001] Divisional application statement

[0002] This application is a divisional application of the Chinese invention patent application with the invention name “Visual Detection and Visual Training Equipment” and application number 202110022647.4 filed on January 8, 2021. Technical Field

[0003] The present invention relates to the field of ophthalmic medical instruments, and in particular to binocular vision detection and vision training equipment. Background Art

[0004] Binocular vision is divided into three levels: the first level is simultaneous vision, the second level is fusional vision (also called "fusion function" or "planar fusion"), and the third level is stereoscopic vision (also called "stereoscopic vision"). Simultaneous vision refers to the signals transmitted from the retinas of both eyes to the brain, which are not received alternately by the two eyes, but simultaneously. Fusional vision is based on binocular simultaneous vision, and the brain center integrates the images falling on the retinas of both eyes into a complete image. Stereoscopic vision is a relatively independent visual function with three-dimensional space based on the above two levels of visual function. It is a higher-level physiological function of binocular adjustment and aggregation under the command of the visual center. For rehabilitation training of patients with eye vision diseases, the ultimate ideal goal is to restore normal binocular vision on the basis of restoring monocular vision.

[0005] Taking patients with amblyopia as an example, the commonly used traditional treatment methods are:

[0006] Patch therapy, which typically involves patching the dominant eye and using only the amblyopic eye, has the disadvantages of potentially reducing vision in the dominant eye and preventing binocular vision training.

[0007] Suppression therapy uses drugs or a translucent membrane to reduce the visual acuity of the dominant eye, thereby suppressing it and restricting its use, forcing the amblyopic eye to use it. However, its disadvantage is that it is difficult to control the degree of suppression. If the suppression level is too low, the amblyopic eye will have difficulty recovering its vision. If the suppression level is too high, the dominant eye's vision may decline. Furthermore, it is impossible to determine whether binocular vision is being trained simultaneously.

[0008] Vision training using a synoptophore involves training binocular vision functions, such as simultaneous vision training, fusion vision training, and stereoscopic vision training. However, its disadvantages are that the instrument is complex to operate, the training content consists of relatively simple static images, and it is less interesting. Patients, especially children, have difficulty maintaining long-term focus, and it is difficult to objectively and quantitatively assess the patient's binocular vision training status. Summary of the Invention

[0009] The purpose of the present invention is to provide a device that can train monocular vision and binocular vision at the same time.

[0010] In order to achieve the above-mentioned purpose, the technical solution of the present invention is to provide a head-mounted visual detection and visual training device, which is a head-mounted device, characterized in that it includes:

[0011] Left eye visual axis measurement module, which can measure the visual axis of the left eye;

[0012] Right eye visual axis measurement module, which can measure the visual axis of the right eye;

[0013] Suppression module: Amblyopia patients have one eye, the amblyopic eye, and one dominant eye. The suppression module suppresses the lens in front of the dominant eye to blur or darken the image. The degree of suppression can be adjusted by replacing lenses with different suppression levels or attaching suppression films with different suppression levels to the transparent lens.

[0014] The ranging module can measure the distance between the human eye and the object in front of the field of view;

[0015] The eye movement point analysis module uses binocular eye movement data to determine binocular fusion vision function and / or binocular stereo vision function in the following ways:

[0016] The method for determining the binocular fusion vision function is as follows: assuming that the distance between the human eye and the viewed two-dimensional image measured by the distance measurement module is Z3, on a plane at a distance Z3 directly in front of the visual field, the intersection of the left eye visual axis and this plane is calculated as the left eye eye movement point, and the intersection of the right eye visual axis and this plane is calculated as the right eye eye movement point, and a distance threshold S5 is set. When the distance between the left eye eye movement point and the right eye eye movement point is less than S5, it is determined that the binocular fusion vision function is in a normal state;

[0017] Among them, the method for judging the binocular stereoscopic vision function is: let the distance between the human eye and the three-dimensional object measured by the ranging module be Z4, let the intersection of the left eye visual axis and the right eye visual axis be the three-dimensional eye movement point, let the distance between the three-dimensional eye movement point and the human eye be Z5, set the distance threshold S6, and when |Z5-Z4|<S6, it is judged that the binocular stereoscopic vision function is in a normal state.

[0018] The present invention has the beneficial effect of simultaneously training both monocular and binocular vision, allowing doctors to objectively assess the status and effectiveness of binocular vision training through eye movement data. Furthermore, training content can be in various forms, such as images, videos, animations, and games, and can be interactive through visual interaction, making training more engaging and encouraging patients to proactively train, thus achieving better visual training results. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the location of each component of the equipment;

[0020] Figure 2(a) is a clear image;

[0021] Figure 2(b) is the image after suppression;

[0022] Figure 3 This is a schematic diagram of binocular stereoscopic vision function judgment. DETAILED DESCRIPTION

[0023] Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiments are only used to illustrate the present invention and are not used in limiting the scope of the present invention.In addition, should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall equally within the scope limited by the appended claims of the application.

[0024] Example 1

[0025] like Figure 1 As shown, the vision testing and training device disclosed in this embodiment includes a display module, an image capture and processing module 105, a calibration module, an eye movement point calculation module, and an eye movement point analysis module. It also includes a computer. The image processing algorithm, calibration module, eye movement point calculation module, and eye movement point analysis module in the image capture and processing module 105 are all programs running on this computer. To further improve the clarity of images captured by the image capture and processing module 105 and reduce eye movement test errors caused by head movement, this embodiment uses a head fixation bracket 104 for head fixation.

[0026] In this embodiment, the display module comprises a polarization display 101 that emits two polarized light patterns, and two polarizers: a left polarizer 102 and a right polarizer 103. The left polarizer 102 is positioned between the left eye and the polarization display 101, while the right polarizer 103 is positioned between the right eye and the polarization display 101. The polarization display 101 can emit horizontally linearly polarized light and / or vertically linearly polarized light. The left polarizer 102 transmits only horizontally linearly polarized light, while the right polarizer 103 transmits only vertically linearly polarized light. Therefore, when the polarization display 101 emits an image formed by horizontally linearly polarized light, only the left eye can see it; when the polarization display 101 emits an image formed by vertically linearly polarized light, only the right eye can see it. Both the left polarizer 102 and the right polarizer 103 transmit near-infrared light. The two polarizers can be fixed to a head-mounted bracket 104 or designed to be worn on the head like glasses. In this embodiment, they are fixed to the head-mounted bracket 104.

[0027] In this embodiment, the image capture and processing module 105 includes a near-infrared camera and two LED near-infrared light sources with an 850nm wavelength. The two near-infrared light sources are located on either side of the near-infrared camera and are point light sources. The near-infrared camera and the near-infrared light sources are placed within the housing of the image capture and processing module 105. The image capture and processing module 105 is located below the polarization display 101. Because both the left polarizer 102 and the right polarizer 103 are transmissive to near-infrared light, the near-infrared camera in the image capture and processing module 105 can capture images of a person's left and right eyes. The near-infrared light source provides illumination for the near-infrared camera and generates corneal reflection points through reflection from the outer surface of the cornea, which can be captured by the near-infrared camera and used as reference points for eye movement point calculation. The image capture and processing module 105 can perform image processing and calculations on the captured eye images to obtain the coordinates of the left pupil center, the center coordinates of each corneal reflection point on the left eye, the center coordinates of the right pupil center, and the center coordinates of each corneal reflection point on the right eye. The image processing algorithm used in this embodiment is as follows: Because the corneal reflection points captured by a near-infrared camera are bright and have a grayscale level of up to 255, two corneal reflection points appear in pairs and are close together. Based on this characteristic, the image regions of the left and right eyes can be found within the entire image captured by the infrared camera. A grayscale threshold is set that is higher than the pupil's grayscale but lower than the grayscale of the surrounding iris and skin areas. Areas below this grayscale threshold are marked as possible pupil areas. An area threshold is then set to eliminate interference from small black objects such as eyelashes, thereby accurately determining the pupil area. The coordinates of the left eye's pupil center are obtained based on the center of the left eye's pupil area. The coordinates of the left eye's corneal reflection point center are obtained based on the average coordinates of the two corneal reflection point centers on the left eye. The left eye's pupil-corneal vector is obtained by subtracting the left eye's corneal reflection point center coordinate from the left eye's pupil center coordinate. Similarly, the coordinates of the right eye's pupil center, the coordinates of the centers of each corneal reflection point on the right eye, and the right eye's pupil-corneal vector can be obtained.

[0028] The function of the calibration module is to calculate the calibration mapping function of the left eye and the calibration mapping function of the right eye by letting the left and right eyes look at several calibration points on the display device respectively.

[0029] The function of the eye movement point calculation module is to calculate the left eye eye movement point coordinates through the left eye pupil cornea vector and the left eye calibration mapping function after calibration is completed; and to calculate the right eye eye movement point coordinates through the right eye pupil cornea vector and the right eye calibration mapping function.

[0030] The function of the eye movement point analysis module is to analyze whether the binocular simultaneous vision function, binocular fusion vision function, and binocular stereoscopic vision function are normal during visual detection and visual training through binocular eye movement data.

[0031] A child patient has amblyopia in the left eye and dominant eye in the right eye. Taking the patient's visual testing and visual training as an example, the specific steps are as follows:

[0032] (1) The patient should first wear glasses of appropriate power to correct refractive error. If there is a large angle of strabismus, prisms can be worn. The patient sits in front of the test device, resting their chin on the head mount 104 and facing the polarization display 101. Both eyes are 60 centimeters away from the polarization display 101. The near-infrared camera in the image capture and processing module 105 continuously captures images covering the bilateral area and calculates the pupil center and corneal reflection point center in real time to obtain the left and right pupil-corneal vectors.

[0033] (2) Calibration

[0034] The eye movement point is the intersection of the eye's visual axis and the display plane. The left eye eye movement point is the intersection of the left eye's visual axis and the display plane, and the right eye eye movement point is the intersection of the right eye's visual axis and the display plane.

[0035] There are two methods for calculating the eye movement point. The first one is to use a single camera system and obtain the mapping relationship between the pupil cornea vector (two-dimensional) in the captured image and the eye movement point coordinates (two-dimensional) on the display plane through multi-point calibration, that is, the mapping function. The second method is to use a binocular camera system through one-point calibration to directly calculate the three-dimensional visual axis, and then calculate the intersection of the visual axis and the display plane to obtain the eye movement point coordinates.

[0036] This embodiment takes the first method as an example.

[0037] The calibration target is displayed in N different positions on the display in sequence, where 2≤N≤9; the pupil-corneal vector and the calibration target coordinates when the left eye looks at the calibration target are substituted into the calibration mapping function equation group, and the left eye calibration mapping function coefficient is solved to obtain the left eye calibration mapping function; the pupil-corneal vector and the calibration target coordinates when the right eye looks at the calibration target are substituted into the calibration mapping function equation group, and the right eye calibration mapping function coefficient is solved to obtain the right eye calibration mapping function. The calibration module calibrates the left eye and the right eye separately, and the order of calibration of the left and right eyes is not limited. In this embodiment, taking 9-point calibration as an example, the calibration targets are 9 points at the center, left, right, top, bottom, upper left, upper right, lower left, and lower right of the display, and the positions of these 9 points are known and determined.

[0038] When calibrating the left eye, the polarization display 101 only emits linearly polarized light in the horizontal direction, so only the left eye can see the calibration point, and the right eye cannot see the calibration point. When calibrating the right eye, the polarization display 101 only emits linearly polarized light in the vertical direction, so only the right eye can see the calibration point, and the left eye cannot see the calibration point. The purpose of calibrating the left and right eyes separately is that the visual acuity of the dominant eye and the amblyopic eye of amblyopic patients is often quite different. If both eyes look at the calibration point at the same time, the visual acuity of the amblyopic eye may be suppressed, and an accurate calibration result cannot be obtained. In addition, considering that the visual acuity of the amblyopic eye is poor, a calibration target that is larger than the calibration target of the dominant eye can be used during calibration.

[0039] The following takes the 9-point calibration process of the left eye as an example.

[0040] Let x s is the horizontal coordinate of the eye movement point on the display plane, y s is the vertical coordinate of the eye movement point on the display plane; e is the horizontal value of the pupil cornea vector, y e is the vertical value of the pupil cornea vector.

[0041] Use the following mapping function①:

[0042]

[0043] The 12 values ​​of a0, a1, a2, a3, a4, a5, b0, b1, b2, b3, b4, and b5 are unknown before calibration. The first stage of calibration is the process of solving these 12 unknowns.

[0044] Because the coordinates of the 9 calibration sight marks on the display plane (x s1 ,y s1 )、(x s2 ,y s2 )、(x s3 ,y s3 )、(x s4 ,y s4 )、(x s5 ,y s5 )、(x s6 ,y s6 )、(x s7 ,y s7 )、(x s8 ,y s8 )、(x s9 ,y s9 ) is known; by the image capture and processing module 105 can be calculated, see these 9 calibration sight mark pupil cornea vectors are (x e1 ,y e1 )、(x e2 ,y e2 )、(xe3 ,y e3 )、(x e4 ,y e4 )、(x e5 ,y e5 )、(x e6 ,y e6 )、(x e7 ,y e7 )、(x e8 ,y e8 )、(x e9 ,y e9 Substituting into mapping function ①, we can obtain the following equation group ② consisting of 18 equations:

[0045]

[0046]

[0047] Because the number of equations is greater than the number of unknown variables, the overdetermined system of equations needs to be solved by the least squares method to obtain the least squares solution a0, a1, a2, a3, a4, a5, b0, b1, b2, b3, b4, b5.

[0048] Since a0, a1, a2, a3, a4, a5, b0, b1, b2, b3, b4, b5 have all been solved to known values, the horizontal direction value x of the pupil cornea vector obtained by the image capture and processing module 105 is e and the vertical value y of the pupil cornea vector e Substituting into mapping function ①, we can find the horizontal coordinate x of the eye movement point on the display plane. s and the vertical coordinate y of the eye movement point on the display plane s The pupil-cornea vector is obtained from the camera image and its unit is pixel.

[0049] After calibrating the left and right eyes using the above steps, the calibration mapping functions for the left and right eyes are calculated. The eye movement point calculation module substitutes the left eye pupil-cornea vector obtained by the image capture and processing module 105 into the left eye calibration mapping function to calculate the eye movement point coordinates for the left eye. The eye movement point coordinates for the right eye are substituted into the right eye calibration mapping function based on the right eye pupil-cornea vector obtained by the image capture and processing module 105.

[0050] (3) Binocular vision balance area suppression test

[0051] Binocular vision training should be conducted when the vision of both eyes is relatively balanced. Because the monocular vision of the amblyopic eye is often significantly different from that of the dominant eye, the purpose of binocular vision balance area measurement is to suppress the image viewed by the dominant eye to a certain extent, so that the vision of the amblyopic eye and the dominant eye is in a relatively balanced state, which is likely to achieve effective binocular vision training results. Suppression is achieved by blurring the entire image or part of the image viewed by the dominant eye, reducing the display resolution, dimming the brightness, or reducing the contrast. The degree of suppression is adjustable, and the image obtained after suppression is called a suppressed image.

[0052] In this embodiment, the suppression method is to perform Gaussian blurring on the image. For example, FIG2( a ) is a clear original image before Gaussian blurring, which is a portion of a vision chart.

[0053] Gaussian blur is a method of performing convolution on an image to blur it. The convolution kernel used in the convolution operation is normally distributed, and its expression is as follows:

[0054]

[0055] Where u and v are the coordinates relative to the center of the convolution kernel. σ is the standard deviation of the convolution kernel's normal distribution. The higher σ is, the greater the blur. Therefore, the degree of blur can be adjusted by setting the value of σ.

[0056] Performing convolution operation on the original image yields the suppressed image after Gaussian blurring, as shown in Figure 2(b).

[0057] In this way, the amblyopic eye sees a clear image, while the dominant eye sees a suppressed image. However, because the amblyopic eye has lower vision, even if the image itself is clear, the patient's subjective perception of the image through the amblyopic eye is still blurry. The following measurement methods can be used to adjust the degree of suppression in the dominant eye to achieve a more balanced vision in both eyes.

[0058] Measurement method 1: The polarized display 101 displays an eye chart visible only to the left eye. Because the patient's left eye is amblyopic, even with refractive correction glasses, the measured monocular visual acuity is only 4.5. The polarized display 101 then displays an eye chart visible only to the right eye. The right eye is the dominant eye, and the measured monocular visual acuity is 5.2. By adjusting the degree of suppression of the eye chart image viewed by the right eye, when the visual acuity corresponding to the eye chart that the right eye can clearly see is 4.3, the suppression degree at this time is the upper limit of suppression; when the visual acuity corresponding to the eye chart that the right eye can clearly see is 4.7, the suppression degree at this time is the lower limit of suppression; when the visual acuity corresponding to the eye chart that the right eye can clearly see is 4.5, the suppression degree reaches the binocular visual equilibrium point. The suppression degree at this time is used as the suppression equilibrium point.

[0059] Measurement method 2: First, the display module displays an optotype A visible only to the left eye (amblyopic eye), and the user is asked to look at optotype A; then, an optotype B visible only to the right eye (dominant eye) is displayed at a position different from optotype A, while optotype A continues to be displayed and its position remains unchanged, and the user is asked to look at optotype B after optotype B appears; a distance threshold S1 is set, the value of S1 is related to the measurement error of the device and the amplitude of the eyeball's own tremor, and the range is generally between 1° and 2°. In this embodiment, S1 = 1.5° visual angle; a time threshold T1 is set, and in this embodiment, T1 = 1 second. If the distance D1 between the user's left eye's eye movement point and optotype A is less than S1 while optotype A is displayed and optotype B is not displayed, and the time when the user's right eye's eye movement point moves to a distance D2 less than S1 from optotype B after optotype B appears is less than T1, then it is judged that the binocular simultaneous vision function is in a normal state. The distance between optotype A and optotype B should not be too small, as this makes it difficult to distinguish eye movement. The distance between optotype A and optotype B should also not be too large, as this reduces visual field sensitivity and makes it difficult to detect optotype B. The distance range can be between 3° and 10° of visual angle. In this embodiment, the distance between optotype A and optotype B is set to 5° of visual angle. The degree of suppression of the dominant eye is adjusted and multiple tests are performed. The interval in which binocular simultaneous vision is determined to be normal according to the above method is selected as the binocular vision balance interval. The highest degree of suppression that maintains both eyes within the binocular vision balance interval is defined as the upper suppression limit, and the lowest degree of suppression that maintains both eyes within the binocular vision balance interval is defined as the lower suppression limit. The suppression balance point is set as the midpoint between the upper and lower suppression limits.

[0060] Measurement Method 3: Optotype E displayed on polarized display 101 is visible only to the left eye (amblyopic eye), and optotype F displayed on polarized display 101 is visible only to the right eye (dominant eye). Optotype E and optotype F are identical in size and shape. Optotype E is displayed at the location of the left eye's eye movement point, while optotype F is displayed at a fixed position set by the program, which in this embodiment is the midpoint of the display plane. The patient uses their eyes to control optotype E so that it overlaps as much as possible. A distance threshold S4 is set. The value of S4 is related to the measurement error of the device and the amplitude of the eye's own tremor, and generally ranges from 1° to 2°. In this embodiment, S4 = 1.5°. When the average distance between optotype E and optotype F remains within a range less than S4, the patient is judged to be in the binocular vision balance range. Adjust the degree of suppression of the dominant eye and perform multiple tests. Select the interval where the average distance between sight marks E and F is less than S4 as the binocular vision balance interval. The highest degree of suppression that maintains both eyes in the binocular vision balance interval is the upper suppression limit, and the lowest degree of suppression that maintains both eyes in the binocular vision balance interval is the lower suppression limit. Set the suppression balance point as the midpoint between the upper and lower suppression limits.

[0061] (4) Binocular vision training

[0062] After measuring the upper and lower limits of suppression, as well as the equilibrium point, through the previous steps, you can select an appropriate suppression level between the upper and lower limits for visual training. The choice of suppression level should be determined based on the patient's specific situation. For example, in the early stages of training for amblyopia, if you want to quickly improve the monocular vision of the amblyopic eye, you can set the suppression level to the upper limit. If, after a period of training, the amblyopic eye's vision has improved to near-normal levels, you can set the suppression level to the equilibrium point.

[0063] There are several ways to conduct visual training:

[0064] Training Method 1: Amblyopia training involves displaying static images, with the left eye (amblyopic eye) viewing a clear image and the right eye (dominant eye) viewing a suppressed image. The clear and suppressed images are displayed in the same position, with the suppression level between the upper and lower suppression limits. In this embodiment, the suppression level is set to the suppression equilibrium point, and the suppression method is Gaussian blur. For example, the training content is a series of brightly colored comic strips. While viewing the images or text, the patient trains the amblyopic eye's vision and simultaneously trains binocular vision. Simultaneously, the eye movement point analysis module calculates in real time whether binocular fusion vision is normal. The distance between the left and right eye movement points is denoted as DLR, and a distance threshold S2 is set. When DLR < S2, binocular fusion vision is considered normal. The value of S2 is related to device measurement error and the amplitude of eye tremor, generally ranging from 1° to 2°. In this embodiment, the distance threshold S2 is set to 1.5°. When DLR < 1.5°, binocular fusion vision is considered normal. If the distance between the left and right eye movement points is ≥1.5°, binocular fusion is considered abnormal, possibly due to visual fatigue, eye muscle fatigue, or brain fatigue. Continuing training may lead to visual abnormalities such as diplopia and visual confusion, which is detrimental to the visual rehabilitation of the amblyopic eye and binocular vision. Training can be paused at this time, and then continued after a period of rest, or the degree of suppression of the dominant eye can be adjusted as appropriate.

[0065] Training Method 2: Amblyopia training involves displaying a dynamic video or animation, where the left eye (the amblyopic eye) sees a clear video or animation, while the right eye (the dominant eye) sees a suppressed video or animation. The clear and suppressed videos or animations are displayed in the same location, and the degree of suppression is between the upper and lower limits. In this embodiment, the suppression level is set to the suppression equilibrium point, and the suppression method is Gaussian blur. For example, the training content is a cartoon that children enjoy watching. While watching the cartoon, the patient trains the vision of the amblyopic eye and simultaneously trains binocular vision. Simultaneously, the eye movement point analysis module calculates in real time whether binocular fusion vision function is normal. In this embodiment, a distance threshold S2 is set to 1.5°. When the distance between the left and right eye movement points is less than 1.5°, binocular fusion vision function is judged to be normal. If the distance between the left and right eye movement points is ≥ 1.5°, binocular fusion vision function is judged to be abnormal. In this case, training can be paused and resumed after a period of rest, or the degree of suppression of the dominant eye can be adjusted as appropriate.

[0066] Training Method 3: The amblyopia training content is an interactive game. Real-time interaction is achieved by controlling the interactive content within the game using eye tracking. The eye tracking point can be the left (amblyopic) eye's eye tracking point, the right (dominant) eye's eye tracking point, or the average of both eyes' eye tracking points. In this embodiment, the eye tracking point of the amblyopic eye is used for interaction. The amblyopic eye sees a clear image, while the dominant eye sees a suppressed image. The clear and suppressed images are displayed at the same location, with the suppression level between the upper and lower limits. In this embodiment, the suppression level is set to the suppression equilibrium point, and the suppression method is Gaussian blur. For example, the training content is a butterfly catching game: butterflies appear from all directions and flutter across the screen. The patient controls a catching net with their eyes. The net's position corresponds to the amblyopic eye's eye tracking point. By focusing on a butterfly for one second, the butterfly is successfully caught. The game interface displays the number of butterflies caught in real time. The game can contain multiple levels. As the levels increase, the number of butterflies fluttering on the screen increases, and their speed increases, increasing the difficulty of the game and the visual requirements. In this embodiment, the distance threshold S2 is set to 1.5°. When the distance between the left eye movement point and the right eye movement point is less than 1.5°, the binocular fusion vision function is judged to be in a normal state. If the distance between the left eye movement point and the right eye movement point is ≥1.5°, the binocular fusion vision function is judged to be in an abnormal state. At this time, the training can be paused and continued after a period of rest, or the degree of suppression of the dominant eye can be adjusted as appropriate. Through the interactive method of the more interesting eye control game, patients, especially children, are more motivated to actively participate, can persist in training for a longer time, and achieve better training results.

[0067] Training method 4: The amblyopia training content is to display a stereoscopic image with binocular parallax, which can be static or dynamic; wherein the left eye (amblyopic eye) sees a clear image, and the right eye (dominant eye) sees a suppressed image, and the degree of suppression is between the upper and lower limits of suppression. In this embodiment, the degree of suppression is selected as the suppression balance point, and the method of suppression is Gaussian blur. For example, the training content is a stereoscopic animation of playing tennis, in which the tennis ball moves back and forth, sometimes far and sometimes near, and the patient is asked to stare at the moving tennis ball with both eyes for visual training. Because the distance of the tennis ball in the stereoscopic animation is pre-set by the program and is known. As Figure 3 As shown, assume that at a certain moment, a stereoscopic image of a tennis ball consists of a left-eye image P1 and a right-eye image P2 with stereoscopic parallax. The distance between its virtual three-dimensional spatial position P3 and the human eye is Z1. Let the intersection of the line connecting the left eye's eye movement points and the line connecting the right eye's eye movement points be P4, and the distance between P4 and the human eye be Z2. Set a distance threshold S3. The value of S3 is related to the device's measurement error and the amplitude of the eye's own tremor. S3 can be a value that changes with the stereoscopic image distance, for example, S3 = Z1 / 10; S3 can also be set to a fixed value, such as 5 cm. In this embodiment, S3 = Z1 / 10. When |Z2-Z1| < S3, the binocular stereoscopic function is judged to be in a normal state. If S3 ≥ Z1 / 10, the binocular stereoscopic function is judged to be in an abnormal state. In this case, the training can be suspended and resumed after a period of rest, or the degree of suppression of the dominant eye can be adjusted as appropriate.

[0068] Training method 5: Amblyopia training includes fixation, saccades, and following, which can generate quantitative eye movement assessment data. By comparing this data with historical data, the training effect can be quantitatively evaluated. This eye movement assessment data includes the amblyopic eye's fixation accuracy, fixation precision, saccade speed, saccade latency, and following accuracy; the dominant eye's fixation accuracy, fixation precision, saccade speed, saccade latency, and following accuracy; and the difference between the two eyes in various indicators. Eye movement assessment can include any one or a combination of these. For example, the training content involves a visual mark located in the center of the display, visible to both the left and right eyes. The patient is required to fixate on the mark with both eyes for at least 10 seconds. The average value of the amblyopic eye's eye movement points and the distance to the mark (fixation accuracy) and the mean square error of the eye movement points (fixation accuracy) during these 10 seconds are then calculated, along with the fixation accuracy and fixation accuracy of the dominant eye's eye movement points during these 10 seconds, and the difference in fixation accuracy and accuracy between the two eyes. The training content can also be a sight mark that randomly jumps and displays at nine azimuth points in the center of the display plane, the left side, the right side, the top side, the bottom side, the upper left corner, the upper right corner, the lower left corner, and the lower right corner. Both the left eye and the right eye can see this sight mark, and the patient is required to scan with both eyes following the position of the sight mark to evaluate the saccadic movement speed and saccadic latency of the amblyopic eye, the saccadic movement speed and saccadic latency of the dominant eye, and the difference in saccadic movement speed and saccadic latency between the two eyes. The training content can also be a sight mark that rotates at a uniform speed along a circular track on the display plane. Both the left eye and the right eye can see this sight mark, and the patient is required to follow the sight mark with both eyes and calculate the average distance between the eye movement points of both eyes and the sight mark to evaluate the following accuracy of the amblyopic eye, the following accuracy of the dominant eye, and the difference in following accuracy between the two eyes.

[0069] To provide real-time insights into what the patient is looking at, these training methods also include a real-time display of eye tracking points. These points can be one or more of the following: the amblyopic eye's point, the dominant eye's point, or the average of both eyes. These points can be displayed as dots, sight marks, or cartoon patterns. These points can be displayed on the monitor the patient is viewing, or on a separate monitor connected to the computer, i.e., a second monitor. This second monitor displays the points and training content in real time, allowing doctors or family members to monitor the training results.

[0070] Example 2

[0071] The display module and image capture and processing module in the present invention may adopt the following methods in addition to the devices designed in Example 1: (one)

[0073] The display module includes a polarization display that can emit circularly polarized light. The polarization display can emit left-handed circularly polarized light or right-handed circularly polarized light. The display module also includes two polarizers, with a left polarizer positioned between the left eye and the polarization display, and a right polarizer positioned between the right eye and the polarization display. The left polarizer is permeable to left-handed circularly polarized light but not right-handed circularly polarized light; the right polarizer is permeable to right-handed circularly polarized light but not left-handed circularly polarized light. When the display device displays an image composed solely of left-handed circularly polarized light, the left eye can see the image through the left polarizer, while the right eye cannot see the image through the right polarizer. When the polarization display displays an image composed solely of right-handed circularly polarized light, the right eye can see the image through the right polarizer, while the left eye cannot see the image through the left polarizer. The image capture and processing module can capture the image for the left eye through the left polarizer and the image for the right eye through the right polarizer. (two)

[0075] The display module is a naked-eye 3D display that can display images visible only to the left eye or only to the right eye. (three)

[0077] The display module includes a display that emits green visible light with a wavelength of 530nm and red visible light with a wavelength of 670nm. The display module also includes two filters: a left filter positioned between the left eye and the display, and a right filter positioned between the right eye and the display. The left filter transmits green light with a wavelength of 530nm but not red light with a wavelength of 670nm; the right filter transmits red light with a wavelength of 670nm but not green light with a wavelength of 530nm. Through the left filter, the left eye can see the image composed of green light with a wavelength of 530nm displayed on the display, but cannot see the image composed of red light with a wavelength of 670nm displayed on the display module. Through the right filter, the right eye can see the image composed of red light with a wavelength of 670nm displayed on the display, but cannot see the image composed of green light with a wavelength of 530nm displayed on the display module. The image capture and processing module can capture the image for the left eye through the left filter and the image for the right eye through the right filter. (Four)

[0079] The display module includes a shutter display device and an automatic shutter lens. The automatic shutter lens includes a left shutter lens and a right shutter lens. The left shutter lens is located between the left eye and the shutter display device, and the right shutter lens is located between the right eye and the shutter display device. When the shutter display device displays an image visible only to the left eye, the left shutter lens is opened and the right shutter lens is closed. At this time, only the left eye can see the image. When the shutter display device displays an image visible only to the right eye, the right shutter lens is opened and the left shutter lens is closed. At this time, only the right eye can see the image. (five)

[0081] The display module is a VR (virtual reality) device. The left eye can only see the image displayed on the left eye display screen of the VR device, and the right eye can only see the image displayed on the right eye display screen of the VR device. The camera in the image capture and processing module is a miniature camera, which shoots at a distance close to the eyes inside the VR. One camera shoots the left eye and the other shoots the right eye. (six)

[0083] The display module is an AR (augmented reality) device. The left eye can only see the image displayed on the left eye display screen of the AR device, and the right eye can only see the image displayed on the right eye display screen of the AR device. The camera in the image capture and processing module is a miniature camera, which shoots at a distance close to the eyes inside the AR. One camera shoots the left eye and the other shoots the right eye.

[0084] Example 3

[0085] In addition to visual training in fixed places such as hospitals and homes, a head-mounted visual detection and visual training device can be used to allow patients to conduct visual training and effectively monitor binocular vision status in their actual daily lives. It includes:

[0086] The left eye visual axis measurement module measures the visual axis of the left eye. In this embodiment, two miniature infrared cameras and two near-infrared LEDs are mounted on the frame to capture the left eye. Using binocular vision principles, the module measures the positions of the two near-infrared corneal reflection points and calculates the three-dimensional position of the center of the left corneal sphere. It also measures the three-dimensional position of the left pupil center. The line connecting the three-dimensional positions of the corneal sphere center and the pupil center is the optical axis of the left eye. The visual axis of the left eye can then be measured after calibration and correction of the Kappa angle.

[0087] The right eye visual axis measurement module measures the visual axis of the right eye. In this embodiment, two miniature infrared cameras and two near-infrared LEDs are mounted on the frame to capture the right eye. The measurement principle is the same as that for the left eye.

[0088] The suppression module applies suppression to the lens in front of the dominant eye, blurring or darkening the image. The degree of suppression can be adjusted by replacing lenses with different suppression levels or applying a suppression film with different suppression levels to the transparent lens. The suppression level is within the range of binocular vision balance.

[0089] The ranging module can measure the distance between the human eye and the object directly in front of the field of view. It can use binocular camera ranging, ultrasonic ranging, laser ranging, etc.

[0090] The eye movement point analysis module uses binocular eye movement data to determine binocular fusion vision function and / or binocular stereo vision function in the following ways:

[0091] The method for determining binocular fusion vision function is to set the distance between the human eye and the two-dimensional image measured by the distance measurement module as Z3. On a plane directly in front of the field of view at a distance of Z3, the intersection of the left eye's visual axis and this plane is calculated as the left eye's eye movement point, and the intersection of the right eye's visual axis and this plane is calculated as the right eye's eye movement point. A distance threshold S5 is set. When the distance between the left eye's eye movement point and the right eye's eye movement point is less than S5, the binocular fusion vision function is judged to be in a normal state. The value of S5 is related to the measurement error of the device and the amplitude of the eyeball's own tremor. For example, a patient wears this device to read text in a book for visual training. The distance measurement module measures the distance between the book and the eyes to be 35cm, that is, Z3 = 35cm. The left-eye visual axis measurement module calculates the intersection of the left-eye visual axis and a plane 35 cm from the eye as the left eye's eye movement point on the book. The right-eye visual axis measurement module calculates the intersection of the right-eye visual axis and a plane 35 cm from the eye as the right eye's eye movement point on the book. In this embodiment, S5 is set to 2 cm. When the distance between the left-eye eye movement point and the right-eye eye movement point is less than 2 cm, the binocular fusion function is judged to be in a normal state. If the binocular fusion function is judged to be in an abnormal state, the suppression level needs to be adjusted until the binocular fusion function returns to normal, or the training is suspended.

[0092] The method for determining binocular stereoscopic vision is to set the distance between the human eye and the three-dimensional object measured by the distance measurement module as Z4, the intersection of the left and right eye visual axes as the three-dimensional eye movement point, the distance between the three-dimensional eye movement point and the human eye as Z5, and set a distance threshold S6. When |Z5-Z4| < S6, binocular stereoscopic vision is determined to be normal. The value of S6 is related to the device's measurement error and the amplitude of the eye's own tremor. For example, a patient wearing this device can be trained by looking at a small ball moving back and forth directly in front of them. At a certain moment, the distance between the human eye and the ball measured by the distance measurement module is 50 cm, i.e., Z4 = 50 cm. The left eye visual axis measurement module calculates the left eye visual axis, and the right eye visual axis measurement module calculates the right eye visual axis. The intersection of the left and right eye visual axes is the three-dimensional eye movement point. In this embodiment, the distance threshold S6 is set to 5 cm. When |Z5-Z4| < 5 cm, binocular stereoscopic vision is determined to be normal. If the binocular stereoscopic vision function is judged to be in an abnormal state, the suppression level needs to be adjusted until the binocular stereoscopic vision function returns to normal, or the training needs to be suspended.

Claims

1. A head-mounted visual inspection and visual training device, which is a head-mounted device, characterized in that: include, Left eye visual axis measurement module, which can measure the visual axis of the left eye; Right eye visual axis measurement module, which can measure the visual axis of the right eye; Suppression module: Amblyopia patients have one eye, the amblyopic eye, and one dominant eye. The suppression module suppresses the lens in front of the dominant eye to blur or darken the image. The degree of suppression can be adjusted by replacing lenses with different suppression levels or attaching suppression films with different suppression levels to the transparent lens. The ranging module can measure the distance between the human eye and the object in front of the field of view; The eye movement point analysis module uses binocular eye movement data to determine binocular fusion vision function and / or binocular stereo vision function in the following ways: The method for determining the binocular fusion vision function is as follows: assuming that the distance between the human eye and the viewed two-dimensional image measured by the distance measurement module is Z3, on a plane at a distance Z3 directly in front of the visual field, the intersection of the left eye visual axis and this plane is calculated as the left eye eye movement point, and the intersection of the right eye visual axis and this plane is calculated as the right eye eye movement point, and a distance threshold S5 is set. When the distance between the left eye eye movement point and the right eye eye movement point is less than S5, it is determined that the binocular fusion vision function is in a normal state; Among them, the method for judging the binocular stereoscopic vision function is: let the distance between the human eye and the three-dimensional object measured by the ranging module be Z4, let the intersection of the left eye visual axis and the right eye visual axis be the three-dimensional eye movement point, let the distance between the three-dimensional eye movement point and the human eye be Z5, set the distance threshold S6, and when |Z5-Z4|<S6, it is judged that the binocular stereoscopic vision function is in a normal state.

Images