Multi-virtual image generation system based on dual-channel optical path multiplexing and ciliary muscle training device

Abstract

The invention discloses a multi-virtual image generation system based on dual-channel optical path multiplexing and a ciliary muscle training device, and relates to the technical field of optical engineering and ophthalmology medical instruments. The device comprises a near-far field double-image generation unit, a light beam combination element, a focal power modulation element and a precise control module, double virtual images with preset diopter difference are generated by optimizing optical parameter design, electronic switching logic and a dynamic positioning mechanism, crystalline lenses are forced to generate physical deformation, and physiologically effective ciliary muscle training is achieved. The light path folding and polarization / amplitude segmentation technology is adopted, the light rays of the two independent image sources are projected to the same visual axis of the human eyes, far and near alternate adjustment stimulation is formed, the training efficiency is improved, and the service life of the device is prolonged.

Description

Technical Field

[0001] This invention belongs to the field of optical engineering and ophthalmic medical device technology, and in particular relates to a multi-virtual image generation system and ciliary muscle training device based on dual-channel optical path multiplexing. Specifically, by using optical path folding and polarization / amplitude segmentation technology, the light from two independent image sources is projected onto the same visual axis of the human eye to form an alternating near and far accommodation stimulus. Background Technology

[0002] In modern society, the widespread use of electronic devices leads to prolonged close-range eye use, causing the ciliary muscles to remain under constant tension. This results in an increasing number of people suffering from accommodative dysfunction, especially teenagers. If accommodative dysfunction is not addressed in time, it may further develop into eye problems such as myopia and eye strain syndrome, seriously affecting visual health.

[0003] Currently, the most commonly used ciliary muscle training tool in clinical practice is the "flippers." Their working principle involves mechanically flipping the positive and negative lenses to change the optical power of the incident light, forcing the ciliary muscle to contract and relax, thereby achieving the training purpose. However, this type of tool has significant drawbacks: First, the mechanical flipping has a significant delay (usually hundreds of milliseconds), making rapid optical power switching impossible and resulting in low training efficiency; second, adjustment parameters (such as refractive error and switching frequency) cannot be quantitatively controlled, leading to varying training effects from person to person and making standardization difficult; third, the mechanical structure is prone to wear after long-term use, causing lens positioning deviations and affecting training accuracy; and fourth, it relies on manual operation by the user, resulting in poor user compliance.

[0004] Existing virtual reality (VR) headsets have been attempted for ciliary muscle training, but these devices employ a fixed focal plane design, and their virtual depth is achieved solely through binocular parallax. When the human eye views the image, the lens's accommodation remains constant, and depth is perceived only through changes in binocular convergence, creating a conflict between accommodation and convergence (VAC phenomenon). This conflict not only fails to achieve effective ciliary muscle training but also exacerbates eye strain and may even induce or worsen myopia, failing to meet the physiological needs of clinical training.

[0005] Therefore, there is an urgent need to develop an optical system that is purely electronically controlled, has no mechanical delay, can generate realistic optical power jumps, and takes into account both device miniaturization and stability, in order to achieve precise, efficient, and comfortable ciliary muscle training and solve many shortcomings of existing technologies. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-virtual image generation system and ciliary muscle training device based on dual-channel optical path multiplexing. Through optical architecture design, electronic control logic and mechanical structure optimization, it solves the problems of mechanical delay, VAC phenomenon and inaccurate positioning of existing training devices, realizes pure electronic, microsecond-level and highly accurate ciliary muscle training, and meets the needs of clinical and home use scenarios, improving training effect and user compliance.

[0007] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0008] This invention relates to a multi-virtual image generation system based on dual-channel optical path multiplexing, comprising the following components arranged sequentially along the optical axis, wherein the optical axes of each component are collinear:

[0009] The first image generation unit (S1) generates a near-field target, and the emitted light rays, after being guided, form a near-field virtual image (V1); the equivalent imaging distance of the near-field virtual image (V1) is... Corresponding regulatory stimulus ;

[0010] The second image generation unit (S2) generates a distant target; the emitted light rays are folded and collimated through an optical path to form a distant virtual image (V2); the equivalent imaging distance of the distant virtual image (V2) is... Corresponding regulatory stimulus ;

[0011] The element combination element (H) is fixed at a 45° angle and uses a semi-transparent semi-reflective mirror or polarizing beam splitter (PBS) to fuse the optical paths of the first image generation unit (S1) and the second image generation unit (S2) so that the two virtual images are projected onto the same visual axis of the human eye.

[0012] The optical power modulation element (M) is disposed on the optical path of the second image generation unit (S2), and is a concave mirror with positive optical power, used to collimate the light emitted by the second image generation unit (S2).

[0013] The first image generation unit (S1) and the second image generation unit (S2) are alternately lit by means of pure electronic means.

[0014] As a preferred technical solution, the distance between the second image generation unit (S2) and the optical power modulation element (M) is... ,satisfy The formula for calculating the image distance D2 of the distant virtual image (V2) is as follows:

[0015] ;

[0016] In the formula, D1 is the object distance between the second image generation unit (S2) and the optical power modulation element (M), F is the focal length of the optical power modulation element (M), and D2 is the distance from V2 to the optical power modulation element (M).

[0017] As a preferred technical solution, the first image generation unit (S1) and the second image generation unit (S2) are selected in the same way, both being high-brightness OLED, Mini-LED or high-definition LCD screens, and the screen parameters meet the following requirements: brightness ≥ 500 cd / m², refresh rate ≥ 120 Hz, resolution ≥ 1080p, pixel density (PPI) ≥ 400, color gamut coverage ≥ sRGB 90%, contrast ratio ≥ 1000:1; the optical parameters of the beam combining element (H) meet the following requirements: working wavelength coverage 400nm-760nm, surface accuracy ≥ λ / 4, substrate material is K9 glass or fused silica; if it is a semi-transparent mirror, the reflectivity / transmittance ratio is 50:50; if it is a polarization beam splitter (PBS), the reflectivity for a specific polarization state is ≥ 90%, and the transmittance is ≥ 90%.

[0018] As a preferred technical solution, the control module uses an ARM chip or an FPGA chip, and establishes communication connections with the first image generation unit (S1) and the second image generation unit (S2) respectively through a MIPI interface; the control module generates sinusoidal control signals of different periods to realize the alternating illumination of the two image generation units; wherein, the expression of the control signal of the first image generation unit (S1) is: The control signal expression for the second image generation unit (S2) is as follows: In the formula, The peak voltage of the signal. To switch frequencies, the adjustable range is: , For time.

[0019] As a preferred technical solution, the specific parameter of the optical power modulation element (M) is the radius of curvature. ,caliber Surface coated with an enhanced silver film or protective aluminum film, reflectivity For medical-grade applications, freeform mirrors are used, with a reflectivity of [missing information]. To eliminate aberrations.

[0020] As a preferred technical solution, a weak negative lens can be selectively connected in series in the optical path of the first image generation unit (S1), through the formula Adjust the equivalent refractive power of V1 To increase the refractive power difference between V1 and V2; where, V1 represents the refractive power of V1 without the series weak negative lens. The focal length (in meters) of a weak negative lens, with a range of values ​​of [value missing]. .

[0021] This invention is a ciliary muscle training device based on dual-channel optical path multiplexing, and also includes a human eye positioning module, a mechanical structure module, an interaction module and a virtual image position verification module;

[0022] The human eye positioning module includes an infrared camera, an infrared LED supplementary light unit, and a positioning algorithm module, which is used to track the pupil position in real time and provide feedback for adjustment.

[0023] The mechanical structure module is used to fix the optical components and the human eye positioning module;

[0024] The interactive module is used by users to select training modes, adjust training parameters, and view training data.

[0025] The virtual image position verification module is used to monitor the actual imaging distance between V1 and V2 to ensure that it meets the preset requirements.

[0026] As a preferred technical solution, the human eye localization module employs a pupil detection algorithm based on Haar features, combined with a dynamic smoothing correction algorithm, using the formula... Real-time correction of pupil center coordinates In the formula, for The pupil center coordinates after constant correction for The coordinates of the pupil center at any given moment. for The constantly detected coordinates of the pupil center This is a smoothing coefficient, with a range of values. ;

[0027] When the pupil deviates from the optical axis threshold When the user deviates beyond a certain point, LED indicators, on-screen arrows, or voice prompts will prompt the user to adjust their posture; Training will be automatically paused when the time is right.

[0028] The human eye positioning module also includes an infrared camera and an infrared LED fill light unit; the infrared camera has a resolution of ≥1280×800, a frame rate of ≥30fps, and a field of view of ≥60°; the infrared LED fill light unit has an emission wavelength of 940nm and a power of ≥50mW to avoid visible light interference.

[0029] As a preferred technical solution, the virtual image position verification module adopts an indirect measurement method, by fixing the focal length. A convex lens images V1 and V2 respectively, and the imaging position d is recorded using an imaging target; the formula is used to... Calculate the actual distance of the virtual image ;

[0030] In the formula, The focal length of the convex lens, with a range of values. , The imaging position recorded for the imaging target. The distance from the virtual image to the convex lens; when Deviation from preset distance When the virtual image position is deemed acceptable.

[0031] As a preferred technical solution, the interaction module includes three training modes: progressive mode, random mode, and custom mode.

[0032] The initial switching frequency of the progressive mode is 0.5Hz, increasing by 0.5Hz every 5 minutes, up to a maximum of 5Hz;

[0033] The switching frequency of the random mode fluctuates randomly within the range of 0.5Hz-5Hz, with a fluctuation period of 1s-3s.

[0034] In the custom mode, users can set the switching frequency (0.5Hz-5Hz), training duration (5min-30min), optotype type (letters, graphics, animation), and refractive error (2.0D-5.0D). The interactive module also includes a display unit and a data interface. The display unit is a ≥2.4-inch LCD or OLED screen used to display real-time training parameters and historical data (training records for the past 30 days). The data interface supports USB or Bluetooth and can export training data to terminal devices. The medical-grade version supports connection to a hospital information system (HIS).

[0035] As a preferred technical solution, the specific configuration of the mechanical structure module is as follows:

[0036] The frame body is made of aluminum alloy or steel with a thickness of ≥3mm and a flatness error of ≤0.01mm.

[0037] Internal optical stage: made of integral casting or CNC machined precision parts, with positioning accuracy ≤ ±0.02mm;

[0038] Contact parts: The headrest, forehead rest, and jaw rest that come into contact with the skin are made of medical-grade polyethylene (PE) or silicone material and are equipped with removable cleaning pads;

[0039] Adjustment structure: The headrest can be adjusted vertically from 3cm to 8cm and forward and backward from 2cm to 5cm, and the position can be locked by a knob or a buckle;

[0040] Housing: Features an anti-reflective coated observation window with a light transmittance of ≥95%, and heat dissipation holes on the side or back with a heat dissipation power of ≥5W.

[0041] The present invention has the following beneficial effects:

[0042] (1) The present invention uses optical path folding and polarization / amplitude segmentation technology to project the light from two independent image sources onto the same visual axis of the human eye, forming an alternating adjustment stimulus between near and far. The dual screens are alternately lit by ARM / FPGA chips, with a switching delay of ≤1μs, which completely eliminates the delay and wear problems of traditional mechanical flipping, and improves training efficiency and equipment lifespan.

[0043] (2) By generating a real optical defocus effect, the refractive power difference between V1 and V2 is ≥2.0D, which forces the lens to undergo physical deformation, avoids the conflict between accommodation and convergence, and achieves physiologically effective ciliary muscle training.

[0044] (3) The present invention uses an infrared camera and a dynamic tracking algorithm to improve the pupil positioning accuracy and ensure that the optical path is precisely aligned with the human eye's visual axis; at the same time, the high rigidity mechanical structure and precision optomechanical stage design reduce the impact of vibration and thermal deformation and improve the stability of optical parameters.

[0045] (4) This invention supports three training modes and multiple parameter customizations to suit users of different ages (children, adults) and different levels of adjustment disorders; the medical-grade version supports data networking and objective detection functions to meet the needs of clinical diagnosis and rehabilitation tracking.

[0046] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This invention relates to a multi-virtual image generation system and ciliary muscle training device based on dual-channel optical path multiplexing.

[0049] Figure 2 This is a schematic diagram of the principle structure of a dual-path optical system;

[0050] Figure 3 A flowchart for human eye positioning. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figure 1-3 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0054] Example 1

[0055] Please see Figures 1-2 As shown, the present invention is a multi-virtual image generation system based on dual-channel optical path multiplexing, comprising the following components arranged sequentially along the optical axis, wherein the optical axes of each component are collinear:

[0056] The first image generation unit (S1) generates a near-field target, and the emitted light rays, after being guided, form a near-field virtual image (V1); the equivalent imaging distance of the near-field virtual image (V1) is... Corresponding regulatory stimulus ;

[0057] The second image generation unit (S2) generates a distant target; the emitted light rays are folded and collimated through the optical path to form a distant virtual image (V2); the equivalent imaging distance of the distant virtual image (V2) is... Corresponding regulatory stimulus ;

[0058] The element combination element (H) is fixed at a 45° angle and uses a semi-transparent semi-reflective mirror or polarizing beam splitter (PBS) to fuse the optical paths of the first image generation unit (S1) and the second image generation unit (S2) so that the two virtual images are projected onto the same visual axis of the human eye.

[0059] The optical power modulation element (M) is disposed on the optical path of the second image generation unit (S2), and is a concave mirror with positive optical power, used to collimate the light emitted by the second image generation unit (S2).

[0060] The first image generation unit (S1) and the second image generation unit (S2) are alternately lit by means of pure electronic means.

[0061] Specifically, the first image generation unit (S1) uses a high-brightness Mini-LED screen with the following parameters: brightness 600 cd / m², refresh rate 180Hz, resolution 1080p, PPI 450, color gamut coverage 95% of sRGB, and contrast ratio 2000:1. Its function is to generate near-field visual targets (such as clear letters or graphics). The emitted light is reflected by the beam combining element (H) and directly enters the human eye, forming a near-field virtual image (V1) with an equivalent imaging distance of 33cm (corresponding to accommodation stimulation +3.0D), providing contraction stimulation to the ciliary muscle.

[0062] The second image generation unit (S2) uses a Mini-LED screen of the same specifications as S1 to generate distant visual targets (such as high-contrast bullseye patterns). The light emitted from it is reflected by the beam combining element (H), refracted towards the optical power modulation element (M), and after being reflected and collimated by M, the light passes through H again and enters the human eye, forming a distant virtual image (V2) with an equivalent imaging distance of 5m (corresponding to an accommodation stimulus of 0.2D), providing relaxation stimulation to the ciliary muscle.

[0063] Beam combining element (H): The selection is based on the application scenario. For consumer-grade scenarios, a semi-transparent mirror is used, while for medical-grade scenarios, a polarizing beam splitter (PBS) is used. In this embodiment, the specific parameters of the consumer-grade version are: reflectivity / transmittance ratio of 50:50, operating wavelength of 400nm-760nm, surface accuracy λ / 4, and substrate material of K9 glass. It is fixed at a 45° angle on the optomechanical stage to ensure precise fusion of the optical paths of S1 and S2, so that the two virtual images are projected onto the same visual axis of the human eye, avoiding ghosting.

[0064] Optical focal length modulation element (M): A concave mirror is selected. The parameters for the consumer-grade version are: focal length F=15cm, radius of curvature 30cm, aperture 5cm, surface coated with enhanced silver film, and reflectivity ≥96%; the medical-grade version uses a freeform mirror with a focal length F=12cm, aperture 6cm, and reflectivity ≥98%, eliminating spherical aberration and coma. The distance D1 between the second image generation unit (S2) and M is 14.5cm (satisfying D1=F-1%F). According to formula (1), the image distance D2 of V2 is calculated to be 4.35m, which is ≤3% of the design target of 5m, meeting the accuracy requirements.

[0065] Control module: Employs an ARM Cortex-A78 chip, integrating two MIPI interfaces to establish communication connections with S1 and S2 respectively. The chip generates a sinusoidal control signal with an adjustable frequency (0.5Hz-5Hz) and a phase difference of 1 / 4 cycle, enabling microsecond-level alternating illumination of the dual screens (switching delay ≤1μs). S2 is completely off when S1 is lit, avoiding crosstalk. In this embodiment, the default switching frequency is 1Hz, and the refractive error difference between V1 and V2 is ≥2.0D, meeting the training stimulus requirements.

[0066] Focal length of the optical power modulation element (M) cm, the distance between the second image generation unit (S2) and the optical power modulation element (M) ,satisfy The formula for calculating the image distance D2 of a distant virtual image (V2) is:

[0067] ;

[0068] In the formula, D1 is the object distance between the second image generation unit (S2) and the optical power modulation element (M), F is the focal length of the optical power modulation element (M), and D2 is the distance from V2 to the optical power modulation element (M).

[0069] The first image generation unit (S1) and the second image generation unit (S2) are selected in the same way, both using high-brightness OLED, Mini-LED or high-definition LCD screens, and the screen parameters meet the following requirements: brightness The optical parameters of the beam combining element (H) must meet the following requirements: refresh rate ≥ 120Hz, resolution ≥ 1080p, pixel density (PPI) ≥ 400, color gamut coverage ≥ 90% of sRGB, and contrast ratio ≥ 1000:1. The working wavelength range of the beam combining element (H) must be 400nm-760nm, surface accuracy ≥ λ / 4, and substrate material must be K9 glass or fused silica. If it is a semi-transparent mirror, the reflectivity / transmittance ratio must be 50:50. If it is a polarization beam splitter (PBS), the reflectivity for a specific polarization state must be ≥ 90%, and the transmittance must be ≥ 90%.

[0070] Specifically, the steps for optimizing lighting parameters are as follows:

[0071] Step 1: Define the core training requirements: Refractive error difference ΔD ≥ 2.0D, V1 equivalent imaging distance 25cm-50cm (ciliary muscle contraction range), V2 equivalent imaging distance 5m-infinity (ciliary muscle relaxation range).

[0072] Step 2, constrain equipment dimensions: The overall length of the desktop equipment is ≤30cm, therefore the focal length F of the optical power modulation element (M) is ≤15cm, so F=15cm is selected;

[0073] Step 3: Determine the relationship between object distance and image distance: Set the distance between S2 and M to D1 = F ± 1% F (to ensure collimation effect), calculate the image distance D2 of V2 using formula (1), and verify whether it is within the range of 5m to infinity;

[0074] Step 4: Optimize the position of S1: Adjust the distance between S1 and H so that the equivalent imaging distance of V1 is within the range of 25cm-50cm, and finally ensure that ΔD≥2.0D;

[0075] Step 5, Iterative Calibration: Detect Dis1 and Dis2 using the virtual image position verification module, and fine-tune the parameters until the deviation is ≤10%. The innovation of this step lies in balancing device miniaturization and virtual image position accuracy through quantitative parameter and formula calculations, thus resolving the technical contradiction of "difficulty in balancing size and performance" in traditional optical path design.

[0076] The control module uses an ARM chip or an FPGA chip and establishes communication connections with the first image generation unit (S1) and the second image generation unit (S2) via a MIPI interface. The control module generates a sinusoidal control signal with a phase difference of 1 / 4 period to achieve alternating illumination of the two image generation units. The expression for the control signal of the first image generation unit (S1) is as follows: The control signal expression for the second image generation unit (S2) is as follows: In the formula, The peak voltage of the signal. To switch frequencies, the adjustable range is: , For time.

[0077] The specific parameters of the optical power modulation element (M) are the radius of curvature. ,caliber Surface coated with an enhanced silver film or protective aluminum film, reflectivity For medical-grade applications, freeform mirrors are used, with a reflectivity of [missing information]. To eliminate aberrations.

[0078] A weak negative lens can be selectively connected in series in the optical path of the first image generation unit (S1), according to the formula. Adjust the equivalent refractive power of V1 To increase the refractive power difference between V1 and V2; where, V1 represents the refractive power of V1 without the series weak negative lens. The focal length (in meters) of a weak negative lens, with a range of values ​​of [value missing]. .

[0079] Example 2

[0080] See Figure 3 As shown, the present invention is a ciliary muscle training device based on dual-channel optical path multiplexing, which can be used to perform the contents of Embodiment 1 and subsequent Embodiment 3 of the present invention, and also includes a human eye positioning module, a mechanical structure module, an interaction module and a virtual image position verification module;

[0081] The human eye positioning module includes an infrared camera, an infrared LED illumination unit, and a positioning algorithm module, which are used to track the pupil position in real time and provide feedback for adjustment.

[0082] The mechanical structure module is used to fix the various optical components and the human eye positioning module;

[0083] The interactive module is used by users to select training modes, adjust training parameters, and view training data;

[0084] The virtual image position verification module is used to monitor the actual imaging distance between V1 and V2 to ensure that it meets the preset requirements.

[0085] The human eye localization module employs a pupil detection algorithm based on Haar features, combined with a dynamic smoothing correction algorithm, using a formula... Real-time correction of pupil center coordinates In the formula, for The pupil center coordinates after constant correction for The coordinates of the pupil center at any given moment. for The constantly detected coordinates of the pupil center This is a smoothing coefficient, with a range of values. ;

[0086] When the pupil deviates from the optical axis threshold When the user deviates beyond a certain point, LED indicators, on-screen arrows, or voice prompts will prompt the user to adjust their posture; Training will be automatically paused when the time is right.

[0087] The optimization steps for pupil dynamic tracking are as follows:

[0088] Step 1, Image Acquisition: The infrared camera acquires images of the human eye at a frame rate of 30fps, and the infrared LED supplementary lighting unit works continuously to ensure clear imaging of the pupil under different ambient light conditions;

[0089] Step 2, Initial Localization: A cascaded classifier based on Haar features detects the pupil region in the image and outputs the preliminary coordinates of the pupil center. ;

[0090] Step 3, Dynamic Correction: Through formula The initial coordinates are smoothed and corrected. When α=0.5, the tracking response speed is guaranteed (delay ≤33ms) and noise interference is effectively suppressed (positioning error ≤0.1mm).

[0091] Step 4, Feedback Control: The corrected pupil center coordinates are... The coordinates are compared with the center coordinates of the optical axis (preset to the image center), the deviation is calculated, and the corresponding feedback mechanism is triggered based on the deviation (prompting adjustment or pausing training). Through a dynamic smoothing correction algorithm, the stability and accuracy of pupil localization are improved, solving the problem that traditional localization algorithms are easily affected by ambient light and micro-movements of the eyeball, ensuring that the optical path is always aligned with the visual axis of the human eye, and guaranteeing the training effect.

[0092] The human eye positioning module also includes an infrared camera and an infrared LED fill light unit; the infrared camera has a resolution of ≥1280×800, a frame rate of ≥30fps, and a field of view of ≥60°; the infrared LED fill light unit has an emission wavelength of 940nm and a power of ≥50mW to avoid visible light interference.

[0093] Specifically, the human eye positioning module consists of a 1280×800 resolution infrared camera (30fps, 65° field of view), a 940nm infrared LED illumination unit (60mW power), and a positioning algorithm module. During operation, the infrared LED illumination unit illuminates the pupil, the camera captures images of the human eye, and a detection algorithm based on Haar features initially locates the pupil center coordinates. Then through the formula ( Dynamic smoothing correction is performed to eliminate positioning fluctuations caused by image noise. When the pupil deviates from the optical axis by ±1mm, a red arrow is displayed at the edge of the screen to prompt the user to adjust their posture; when the deviation exceeds ±2mm, the control module sends a pause signal to stop the switching between the two screens until the user adjusts the position, ensuring that the optical path is always aligned with the visual axis of the human eye.

[0094] Specifically, the steps for electronic switching between the two screens are as follows:

[0095] Step 1, Hardware Setup: Select an ARM chip that supports dual-channel MIPI output and design a high-speed switching circuit to ensure that the chip can independently control the on / off state of S1 and S2.

[0096] Step 2, Signal Design: Generate a sinusoidal control signal with a phase difference of 1 / 4 period. The control signal for S1 is... The control signal for S2 is ( (based on the screen's rated drive voltage) to ensure smooth, flicker-free switching;

[0097] Step 3, Crosstalk Suppression: Set up hardware interlock logic. The blanking interval ensures that the driving voltage of S2 is 0V when S1 is lit, so that it is completely extinguished and optical crosstalk is avoided.

[0098] Step 4, Frequency Adjustment: The chip receives the user-set switching frequency f through a software interface and adjusts the control signal cycle in real time to achieve continuous adjustment from 0.5Hz to 5Hz. It uses purely electronic signals to control the alternating lighting of the two screens, completely eliminating mechanical structures and achieving microsecond-level switching. This solves the delay, wear, and crosstalk problems of traditional mechanical flipping, and the adjustable switching frequency adapts to different training intensity requirements.

[0099] The virtual image position verification module uses an indirect measurement method, by fixing the focal length. A convex lens images V1 and V2 respectively, and the imaging position d is recorded using an imaging target; the formula is used to... Calculate the actual distance of the virtual image ;

[0100] In the formula, The focal length of the convex lens, with a range of values. , The imaging position recorded for the imaging target. The distance from the virtual image to the convex lens; when Deviation from preset distance When the virtual image position is deemed acceptable.

[0101] Specifically, the virtual image position verification module uses a high-precision convex lens with a focal length of f=0.1m, fixed at the human eye's observation position. During verification, S1 and S2 are illuminated respectively, and the imaging positions d1 and d2 of V1 and V2 after passing through the convex lens are recorded through the imaging target. These positions are then substituted into the formula. Calculate the actual distances Dis1 and Dis2. In this embodiment, the preset distance of V1 is 33cm, the measured d1=15.2cm, and the calculated Dis1=30.8cm, with a deviation ≤7%; the preset distance of V2 is 5m, the measured d2=10.2cm, and the calculated Dis2=5.1m, with a deviation ≤2%, both meeting the qualified standard of deviation ≤10%. If the deviation exceeds the standard, the positions of S1, S2, or M are finely adjusted using the multi-dimensional adjustment frame (accuracy ±0.01mm) on the optical machine until the requirements are met.

[0102] The specific steps for verifying the virtual image are as follows:

[0103] Step 1, Equipment Setup: Fix a high-precision convex lens with a focal length of f=0.1m at the human eye's observation position, and place the imaging target (resolution 0.01mm) on the other side of the convex lens. The distance between the target and the convex lens can be adjusted.

[0104] Step 2, Imaging Recording: Illuminate S1 alone, move the imaging target until a clear real image of V1 is observed on the target, and record the distance d1 between the target and the convex lens; similarly, illuminate S2 alone and record the distance d2 corresponding to the real image of V2.

[0105] Step 3: Calculation and Verification: Substitute f, d1, and d2 into the formula. Calculate the actual distances Dis1 and Dis2 between V1 and V2 respectively;

[0106] Step 4, Adjustment and Calibration: If the deviation of Dis from the preset distance is >10%, fine-tune the front-to-back position of S1 (adjustment accuracy ±0.01mm) or the tilt angle of M (adjustment accuracy ±0.01°) using the multi-dimensional adjustment frame on the optical stage. Repeat steps 2-3 until the deviation is ≤10%. The innovation of this step lies in proposing a simple and low-cost indirect measurement method that transforms the virtual image distance, which cannot be directly measured, into a real image position that can be accurately measured. This solves the technical problem of verifying the positioning accuracy of the virtual image and ensures the stable and reliable optical performance of the system.

[0107] The interactive module features a 2.8-inch LCD touchscreen and supports three training modes: progressive mode, random mode, and custom mode.

[0108] The initial switching frequency of progressive mode is 0.5Hz, increasing by 0.5Hz every 5 minutes, up to a maximum of 5Hz, gradually increasing the training intensity, which is suitable for beginners;

[0109] The switching frequency of the random mode fluctuates randomly within the range of 0.5Hz-5Hz (fluctuation period of 2s), which trains the ciliary muscle's rapid reaction ability and is suitable for advanced training.

[0110] In custom mode, users can set the switching frequency (0.5Hz-5Hz), training duration (5min-30min), optotype type (letters, graphics, animation), and refractive error (2.0D-5.0D) to suit individual needs. The interaction module also includes a display unit and a data interface. The display unit is a ≥2.4-inch LCD or OLED screen used to display real-time training parameters and historical data (training records for the past 30 days). The data interface supports USB or Bluetooth and can export training data to terminal devices. The medical-grade version supports connection to the hospital information system (HIS) to facilitate doctors in tracking patients' recovery progress.

[0111] The specific configuration of the mechanical structure module is as follows:

[0112] The main frame is made of aluminum alloy or steel with a thickness of ≥3mm and a flatness error of ≤0.01mm;

[0113] The internal optical stage is made of integral casting or CNC machined precision parts, with a positioning accuracy of ≤±0.02mm;

[0114] The contact parts, such as the headrest, forehead rest, and jaw rest, which come into contact with the skin, are made of medical-grade polyethylene (PE) or silicone material and are equipped with removable cleaning pads.

[0115] The headrest of the adjustable structure has a vertical adjustment range of 3cm-8cm and a front-to-back adjustment range of 2cm-5cm, and the position can be locked by a knob or a buckle.

[0116] The outer casing has an anti-reflective coating viewing window with a light transmittance of ≥95%, and heat dissipation holes on the side or back with a heat dissipation power of ≥5W.

[0117] Specifically, the main frame is made of 5mm thick aluminum alloy sheet, CNC machined with a flatness error of ≤0.01mm; the internal optical stage is a precision-cast component with a positioning accuracy of ≤±0.02mm, used to fix various optical components and reduce the impact of vibration and thermal deformation on the optical path. The headrest, forehead rest, and jaw rest that come into contact with the skin are made of medical-grade silicone material and are equipped with removable cleaning pads; the headrest has a 5cm vertical adjustment range and a 3cm horizontal adjustment range, which can be locked in position via a knob to accommodate users with different head shapes. The outer shell is made of ABS engineering plastic, with an anti-reflective coating observation window (light transmittance ≥96%) at the front, ventilation holes on the sides, and a built-in 5W cooling fan to ensure stable operation of the equipment for extended periods.

[0118] Example 3

[0119] This invention relates to a consumer-grade desktop ciliary muscle training device. The corresponding method is the result of combining the technical solution of Embodiment 1 with a specific platform, tools, and algorithm. Furthermore, this invention describes the basic methodological theory, and the specific implementation process is as follows:

[0120] (1) Component selection and parameter settings are as follows:

[0121] In the multi-virtual-image generation system, S1 and S2 use 1080p Mini-LED screens with high brightness. The system features a refresh rate of 180Hz and a PPI of 450. H is a semi-transparent mirror with a ratio of R:T=50:50, K9 glass substrate, and a surface finish of λ / 4. M is a concave mirror with an F=15cm aperture of 5cm and a reflectivity ≥96%. The control module uses an ARM Cortex-A78 chip with a switching frequency of 0.5Hz-5Hz. Optical parameters include a distance of D1=14.5cm between S2 and M, an equivalent imaging distance of 33cm (+3.0D) for V1, an equivalent imaging distance of 4.35m (0.23D) for V2, and a refractive error of 2.77D.

[0122] The human eye positioning module uses a 1280×800 infrared camera, a 940nm infrared LED, and a dynamic correction algorithm with α=0.5; the mechanical structure module uses an aluminum alloy frame, a medical silicone headrest (adjustable up and down by 5cm, and forward and backward by 3cm), an ABS shell, and an anti-reflective observation window; the interaction module uses a 2.8-inch LCD touchscreen, supports progressive / random / custom modes, and USB data export; the virtual image position verification module uses an f=0.1m convex lens, with an imaging target resolution of 0.01mm.

[0123] (2) The assembly process is as follows:

[0124] Mechanical structure assembly: Fix the aluminum alloy frame on the workbench, install the internal optical stage, and ensure that the flatness error of the optical stage is ≤0.01mm;

[0125] Optical component installation: According to the optical path architecture diagram, fix H (45° tilt), M, S1, and S2 on the optomechanical stage, and calibrate the optical axes of each component using a laser collimator to ensure collinearity (deviation ≤ 0.02 mm).

[0126] Circuit connection: Connect the control module to S1 and S2 via the MIPI interface, connect the infrared camera and infrared LED to the control module, and ensure that the circuit is unobstructed;

[0127] Housing encapsulation: Install ABS housing, viewing window and cooling fan, and fix touch screen and USB interface.

[0128] (3) The calibration procedure is as follows:

[0129] Optical path calibration: Activate the control module, illuminate S1 separately, detect the actual distance Dis1 of V1 through the virtual image position verification module, and adjust the front and rear positions of S1 to achieve the desired optical path. Light up S2 separately, adjust the tilt angle of M, so that... ;

[0130] Switching calibration: Set the switching frequency to 1Hz, and use a high-speed camera (1000fps) to capture the dual-screen switching process to verify that the switching delay is ≤1μs and there is no crosstalk;

[0131] Positioning calibration: Activate the human eye positioning module, simulate the human eye using a standard pupil model (5mm in diameter), move the model, and verify that the positioning algorithm's correction accuracy is ≤0.1mm, with feedback indicating accuracy.

[0132] (4) Training process:

[0133] Power-on preparation: After the user connects the power supply, the device performs a self-test (optical parameters, positioning module, screen status). After the self-test passes, the touch screen displays the main interface.

[0134] Position adjustment: The user wears a headrest and adjusts its position so that the eyes are aligned with the observation window. The touch screen displays the pupil positioning status in real time (green indicates qualified, red indicates deviation).

[0135] Mode selection: Users can select the training mode (such as progressive mode) and set the training duration (such as 15 minutes) via the touch screen.

[0136] Start training: After the device confirms that the pupil positioning is qualified, the training will start automatically. S1 and S2 will light up alternately at the set frequency, and the screen will display the preset target. The user will naturally adjust the lens by following the target.

[0137] Real-time feedback: During training, the touchscreen displays the remaining time, current switching frequency, and pupil deviation. If the pupil deviation exceeds the limit, the screen arrow will prompt for adjustment.

[0138] Training ends: After the set duration is reached, the device will automatically stop training and display the training data (duration, average frequency, refractive error difference, average pupil deviation). Users can choose to export the data or retrain.

[0139] Equipment maintenance: The positions of V1 and V2 are checked weekly using the virtual image position verification module. If the deviation exceeds the standard, adjustments are made according to the calibration procedure.

[0140] It is worth noting that the various units included in the above system embodiments are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0141] Furthermore, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware, and the corresponding program can be stored in a computer-readable storage medium.

[0142] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multi-virtual image generation system based on dual-channel optical path multiplexing, characterized in that, Comprising the following components arranged in order along the optical axis direction, and the optical axes of each component are collinear: The first image generating unit (S1) is used for generating a near view target, and the light rays emitted form a near distance virtual image (V1) after being transmitted; the equivalent imaging distance of the near distance virtual image (V1) is corresponding adjustment stimulus ; The second image generating unit (S2) is used for generating a far vision target, and light rays emitted after light path folding and collimation form a far distance virtual image (V2); an equivalent imaging distance of the far distance virtual image (V2) is corresponding adjustment stimulus ; Element combination element (H) is fixed at 45° inclination, using half-transmission half-reflection mirror or polarization beam splitter, for fusing the light path of the first image generation unit (S1) and the second image generation unit (S2), so that the double virtual images are projected to the same visual axis of the human eye; Power modulation element (M) is arranged on the light path transmission path of the second image generation unit (S2), which is a concave mirror with positive power, for collimating the light emitted by the second image generation unit (S2); The first image generation unit (S1) and the second image generation unit (S2) are alternately lit by pure electronic method; The control module is electrically connected with the first image generation unit (S1) and the second image generation unit (S2) for controlling the pure electronic alternately lighting of the two, realizing the refractive power microsecond level jump, and the refractive power difference between V1 and V2 is greater than or equal to 2.0D.

2. The multi-virtual image generation system based on dual-channel optical path multiplexing according to claim 1, characterized in that, The second image generating unit (S2) is at a distance from the power modulation element (M) , which satisfies ; and a formula for calculating an image distance D2 of the far distance virtual image (V2) is ； In the formula, D1 is the object distance of the second image generation unit (S2) and the power modulation element (M), F is the focal length of the power modulation element (M), and D2 is the distance from V2 to the power modulation element (M).

3. The multi-virtual image generation system based on dual-channel optical path multiplexing according to claim 1, wherein, The first image generation unit (S1) and the second image generation unit (S2) are selected in the same way, both are high-brightness OLED, Mini-LED or high-definition LCD screen, and the screen parameters meet: brightness≥500cd / m², refresh rate≥120Hz, resolution≥1080p, pixel density (PPI)≥400, color gamut coverage≥sRGB 90%, contrast≥1000:1; The optical parameters of the light beam combination element (H) meet: the working waveband covers 400nm-760nm, the surface precision is≥λ / 4, the substrate material is K9 glass or fused quartz; If it is a half-transmission half-reflection mirror, the reflectivity / transmissivity ratio is 50:50; If it is a polarization beam splitter (PBS), the reflectivity of a specific polarization state is≥90%, and the transmissivity is≥90%.

4. The multi-virtual image generation system based on dual-channel optical path multiplexing according to claim 1, wherein, The control module adopts an ARM chip or an FPGA chip, and is in communication connection with a first image generating unit (S1) and a second image generating unit (S2) through an MIPI interface; the control module generates sinusoidal control signals with different periods to realize the alternate lighting of the two image generating units; wherein the control signal expression of the first image generating unit (S1) is: The control signal expression of the second image generating unit (S2) is: In the formula, is a signal peak voltage, is a switching frequency, and the adjustable range is , is time.

5. The multi-virtual image generation system based on dual-channel optical path multiplexing according to claim 1, characterized in that, The specific parameter of the power modulating element (M) is the radius of curvature , aperture , surface plating enhanced silver film or protective aluminum film, reflectivity ; For medical grade applications, a freeform mirror is used, with reflectivity .

6. The multi-virtual image generation system based on dual-channel optical path multiplexing according to claim 1, wherein, A weak negative lens can be selectively connected in series in the light path of the first image generating unit (S1), through the formula Adjust the equivalent refractive power of V1 to increase the refractive power difference between V1 and V2; wherein, V1 is the refractive power of V1 without connecting the weak negative lens in series, f is the focal length of the weak negative lens, and the value range is .

7. A ciliary muscle training device based on dual channel optical path multiplexing, characterized by, The multi-virtual image generation system of any of claims 1-6 further comprises: a human eye positioning module, a mechanical structure module, an interactive module and a virtual image position verification module; The human eye positioning module comprises an infrared camera, an infrared LED light supplement unit and a positioning algorithm module, for real-time tracking of pupil position and feedback adjustment; The mechanical structure module is used for fixing each optical component and the human eye positioning module; The interactive module is used for user to select training mode, adjust training parameters and refer to training data; The virtual image position verification module is used for monitoring the actual imaging distance of V1 and V2 to ensure that the preset requirements are met.

8. The ciliary muscle training device based on dual-channel optical path multiplexing according to claim 7, characterized in that, The human eye localization module employs a pupil detection algorithm based on Haar features, combined with a dynamic smoothing correction algorithm, using the formula... Real-time correction of pupil center coordinates In the formula, for The pupil center coordinates after constant correction for The coordinates of the pupil center at any given moment. for The constantly detected coordinates of the pupil center. This is a smoothing coefficient, with a range of values. ; When the pupil deviates from the optical axis threshold , the user is prompted to adjust the pose by LED indicator lights, screen arrows, or voice; when the deviation exceeds , the training is automatically paused.

9. The ciliary muscle training device based on dual channel optical path multiplexing according to claim 7, wherein, The virtual image position verification module uses an indirect measurement method, by fixing the focal length. A convex lens images V1 and V2 respectively, and the imaging position d is recorded using an imaging target; the formula is used to... Calculate the actual distance of the virtual image ; In the formula, is the focal length of the convex lens, and the value range is , is the imaging position of the imaging target record, is the distance from the virtual image to the convex lens; when the deviation from the preset distance is , it is determined that the virtual image position is qualified.

10. The ciliary muscle training device based on dual channel optical path multiplexing according to claim 7, wherein, The interactive module includes three training modes: gradual mode, random mode and custom mode: The initial switching frequency of the gradual mode is 0.5Hz, which increases by 0.5Hz every 5 minutes, and the highest can reach 5Hz; The switching frequency of the random mode randomly fluctuates in the range of 0.5Hz-5Hz, and the fluctuation period is 1s-3s; In the custom mode, the user can set the switching frequency, training duration, target type and diopter difference; the interaction module further includes a display unit and a data interface, the display unit is a ≥2.4 inch LCD or OLED screen, which is used to display real-time training parameters and historical data; the data interface supports USB or Bluetooth, and the training data can be exported to a terminal device; the medical grade version supports connection with a hospital information system.

Images