Eye behavior monitoring method, smart glasses and system

Abstract

This application discloses a method, smart glasses, and system for monitoring eye use behavior, relating to the field of wearable device technology. Specifically, it is applied to smart glasses, which include a multimodal sensor. The method includes: acquiring eye use behavior data collected by the multimodal sensor; extracting eye use behavior features based on the data, wherein the features include at least two of the following: eye use distance, eye use duration, head posture, and ambient light intensity; assessing the eye health level corresponding to each feature; fusing these features to obtain a comprehensive eye health level; and outputting a prompt if the comprehensive eye health level is less than a preset threshold. This application improves the accuracy and reliability of eye use behavior reminders.

Description

Technical Field

[0001] This application relates to the field of wearable device technology, and in particular to a method for monitoring eye behavior, smart glasses and system. Background Technology

[0002] Currently, with the widespread use of electronic devices and increasing academic pressure, the eye strain on children and adolescents is increasing, and myopia is showing a serious trend of high incidence and younger age of onset. Therefore, effective monitoring and reminders of daily eye use behavior are considered one of the important means of myopia prevention and control.

[0003] Traditional eye behavior monitoring methods typically rely on a single type of sensor to detect a specific eye behavior parameter (such as viewing distance or continuous viewing time) in real time. When the parameter is detected to be outside the preset normal range, the system triggers an alert; once the parameter returns to the normal range, the system automatically clears the current abnormal state and resumes monitoring.

[0004] The aforementioned monitoring methods have significant shortcomings in practical applications, particularly in the accuracy of alerts. Because users' daily eye-use behavior involves numerous instantaneous and accidental changes in movement, the judgment of exceeding thresholds for a single parameter is easily affected by such disturbances, leading to misjudgments and frequent false triggers of alerts or missed alerts when they are truly needed. Therefore, how to improve the accuracy of eye-use behavior alerts to reduce invalid or missed alerts has become an urgent technical problem to be solved in this field. Summary of the Invention

[0005] The main objective of this application is to provide a method, smart glasses and system for monitoring eye use behavior, aiming to solve the technical problem of how to improve the accuracy of eye use behavior reminders in order to reduce invalid or missed reminders.

[0006] To achieve the above objectives, this application provides a method for monitoring eye use behavior, wherein the smart glasses include a multimodal sensor, and the method for monitoring eye use behavior includes the following steps: The eye-use behavior data collected by the multimodal sensor is acquired, and eye-use behavior features are extracted based on the eye-use behavior data. The eye-use behavior features include at least two of the following: eye-use distance, eye-use duration, head posture, and ambient light intensity. Assess the visual health status corresponding to each of the aforementioned visual behavior characteristics, and integrate the visual health status of each of the aforementioned visual health status to obtain a comprehensive visual health status. If the overall eye health level is less than a preset threshold, a prompt message will be output.

[0007] In addition, to achieve the above objectives, this application also provides a smart glasses, which includes a multimodal sensor and a processor; The multimodal sensor is used to collect users' eye behavior data; The processor is used to perform the steps of the eye behavior monitoring method described above.

[0008] In addition, to achieve the above objectives, this application also provides an eye use behavior monitoring system, which includes smart glasses and a cloud server connected in communication, and the smart glasses include a multimodal sensor; The multimodal sensor is used to collect users' eye behavior data; The cloud server is used to perform the steps of the eye behavior monitoring method described above.

[0009] One or more technical solutions proposed in this application have at least the following technical effects: This application embodiment collects at least two different types of eye-use behavior data through multimodal sensors on smart glasses and extracts corresponding eye-use behavior features, such as eye-use distance, eye-use duration, head posture, and ambient light intensity (any two or more of these). Then, it assesses the eye health level corresponding to each feature and merges these multiple health levels to obtain a comprehensive eye health level. A prompt message is only output when this comprehensive level is below a preset threshold. Instead of relying on a single parameter's instantaneous threshold exceeding, it comprehensively considers multiple dimensions of eye-use behavior. Even if a certain dimension experiences a brief abnormal fluctuation due to accidental movement, the normal health levels of other dimensions can offset the abnormal impact to some extent, making the comprehensive eye health level less prone to drastic changes. Therefore, it can effectively filter out instantaneous and accidental movement disturbances, reduce frequent triggering of invalid reminders due to a single parameter's brief exceedance, and prevent premature clearing of abnormal states due to a single parameter temporarily returning to normal, thus avoiding the omission of scenarios that truly require continuous reminders. Ultimately, this improves the accuracy and reliability of eye-use behavior reminders. Attached Figure Description

[0010] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating the first embodiment of the eye use behavior monitoring method of this application; Figure 2This is a schematic diagram of the sensor setup involved in an embodiment of the eye behavior monitoring method of this application; Figure 3 This is a schematic diagram of the virtual avatar generation process in one embodiment of the eye behavior monitoring method of this application; Figure 4 This is a schematic diagram of the system architecture of the smart glasses in this application; Figure 5 This is a schematic diagram of the system architecture of the eye behavior monitoring system of this application.

[0013] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, 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.

[0015] In recent years, myopia among children and adolescents has shown a high incidence and younger age of onset, becoming a major public health issue. Scientific research indicates that poor eye habits are the core factors leading to the occurrence and development of myopia, mainly including: insufficient outdoor activity time, excessive use of electronic products, poor reading and writing posture, excessively close viewing distance, inadequate visual environment illumination, and lack of eye relaxation techniques.

[0016] To address the aforementioned issues, various smart eye-protection glasses products have emerged in the current technology. These products monitor parameters such as viewing distance, head tilt angle, and ambient light by combining multiple sensors, and issue reminders via vibrators or bio-lights. However, existing technologies suffer from at least the following problems and drawbacks: the intervention logic is simplistic and crude. Existing reminders are usually based on a single threshold (e.g., triggered when viewing distance is <30cm), lacking a comprehensive assessment of multi-dimensional behavior, which easily leads to "false positives" or invalid reminders. For children and adolescents with weaker self-control, such passive reminders can easily lead to "reminder fatigue," where children may gradually ignore or even resent these intrusive prompts, resulting in decreased long-term wear compliance and failing to achieve truly continuous improvement in eye-use behavior.

[0017] To address the aforementioned issues, the main solution of this application is as follows: acquiring eye-use behavior data collected by the multimodal sensor, and extracting eye-use behavior features based on the data, wherein the eye-use behavior features include at least two of the following: eye-use distance, eye-use duration, head posture, and ambient light intensity; evaluating the eye health level corresponding to each of the aforementioned eye-use behavior features, and fusing the eye health levels to obtain a comprehensive eye health level; if the comprehensive eye health level is less than a preset threshold, then outputting a prompt message.

[0018] This application collects at least two different types of eye-use behavior data using multimodal sensors on smart glasses and extracts corresponding eye-use behavior features, such as eye-use distance, eye-use duration, head posture, and ambient light intensity (any two or more of these). It then assesses the eye health level corresponding to each feature and merges these multiple health levels to obtain a comprehensive eye health level. A prompt is only output when this comprehensive level falls below a preset threshold. Instead of relying on a single parameter's instantaneous threshold exceedance, it comprehensively considers multiple dimensions of eye-use behavior. Even if a certain dimension experiences a brief abnormal fluctuation due to accidental movement, the normal health levels of other dimensions can offset the abnormal impact to some extent, making it less likely for the comprehensive eye health level to fluctuate drastically. Therefore, it can effectively filter out instantaneous and accidental movement disturbances, reduce frequent triggering of invalid reminders due to a single parameter's brief exceedance, and prevent premature clearing of abnormal states due to a single parameter temporarily returning to normal, thus avoiding the omission of scenarios that truly require continuous reminders. Ultimately, this improves the accuracy and reliability of eye-use behavior reminders.

[0019] It should be noted that the implementing entity of the various embodiments of the eye behavior monitoring method of this application can be a smart glasses capable of realizing the above functions.

[0020] Based on this, this application proposes a method for monitoring eye use behavior according to a first embodiment. In this embodiment, the method for monitoring eye use behavior is applied to smart glasses, as shown below. Figure 1 As shown, the eye use behavior monitoring method includes the following steps S10~S30: Step S10: Obtain eye behavior data collected by the multimodal sensor, and extract eye behavior features based on the eye behavior data, wherein the eye behavior features include at least two of eye distance, eye duration, head posture and ambient light intensity. Multimodal sensors may include time-of-flight sensors, camera sensors, motion sensors (such as inertial measurement units, IMUs), and ambient light sensors. The time-of-flight sensor is used to collect distance data between the user's eyes and the object being viewed in real time; the camera sensor is used to collect images of the user's eyes and the object being viewed to help identify the type of object being viewed (such as a book or an electronic screen) and determine the duration of continuous eye use; the motion sensor is used to capture head posture data such as the tilt angle and pitch state of the user's head; and the ambient light sensor is used to detect the light intensity data of the user's viewing environment.

[0021] Based on the collected eye-use behavior data, corresponding eye-use behavior features can be extracted through feature extraction algorithms. These features include at least two of the following: eye-use distance, eye-use duration, head posture, and ambient light intensity. Each feature reflects different dimensions of the user's eye-use status, providing comprehensive and reliable data support for subsequent health assessments.

[0022] In a preferred embodiment, the eye-use behavior characteristics include four factors: eye-use distance, eye-use duration, head posture, and ambient light intensity, to comprehensively reflect the user's multi-dimensional eye-use behavior and improve the comprehensiveness and accuracy of health assessment. For ease of description, the following embodiments will be illustrated using the above four types of eye-use behavior characteristics as examples.

[0023] Step S20: Assess the visual health level corresponding to each of the aforementioned visual behavior characteristics, and integrate the visual health levels of each to obtain a comprehensive visual health level; Based on eye health guidelines, corresponding health evaluation standards can be set for each eye behavior characteristic in advance. For example, a reasonable range threshold can be set for eye distance, a maximum allowable duration for eye use, a standard angle range for head posture, and a suitable lighting range for ambient light. The actual detection values ​​of each characteristic can be compared with the corresponding evaluation standards to quantify and obtain the eye health score corresponding to each characteristic.

[0024] Then, a preset fusion algorithm is used to combine the influence weights of each eye-use behavior feature on eye health, and the eye health scores corresponding to all features are weighted and fused to calculate the final comprehensive eye health level that can objectively reflect the user's current overall eye use status, thus avoiding the one-sidedness of single feature evaluation.

[0025] In one exemplary implementation, a weighted fusion algorithm is used to calculate the comprehensive eye health score S(t), and the calculation formula is: S(t) = α·fD(d) + β·fT(t) + γ·fP(p) + δ·fL(l). Here, α, β, γ, and δ are dynamic weight parameters that can be adjusted according to the user's age or environmental pattern; fD(d) is a distance risk function, which performs a non-linear mapping based on viewing distance, with more severe deductions for closer distances, and a perfect score for distances greater than 40cm; fT(t) is a duration risk function, which uses an integral accumulation logic, with a linear decay in score for longer periods of close-range viewing of electronic screens; fP(p) is a posture risk function, which calculates head posture using IMU data, deducting points when the pitch angle deviates from the standard value or when there is a head tilting motion; fL(l) is an illuminance risk function, triggering deductions when the ambient light is too dim (less than 300 lux). The value of S(t) is limited to between 0 and 100, with higher values ​​indicating healthier eye health.

[0026] Step S30: If the overall eye health level is less than a preset threshold, a prompt message is output.

[0027] If the overall eye health level is lower than a preset threshold, a prompt message is output. A preset threshold for overall eye health level is set between 0 and 100, for example, a threshold of 50. The current overall eye health score S(t) is calculated in real time and compared with the preset threshold.

[0028] When the overall eye health level is continuously below a preset threshold for a period exceeding the preset anti-shake duration, or when the overall eye health level momentarily falls below the preset threshold, a prompt message is triggered. This prompt message can be at least one of the following: voice prompt, vibration prompt, visual indicator light prompt, or text and icon prompts on the smart glasses lenses, used to remind the user to adjust their eye posture, rest, or improve ambient lighting.

[0029] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, the multimodal sensor includes a time-of-flight sensor, a motion sensor, a camera sensor, and an ambient light sensor. The step of acquiring the eye-use behavior data collected by the multimodal sensor and extracting eye-use behavior features based on the eye-use behavior data includes: Step A10: Obtain the distance data collected by the time-of-flight sensor, and extract the eye distance based on the distance data; A time-of-flight sensor emits modulated infrared light pulses toward an object in front of the user and receives the light signals reflected back from the target. By calculating the time of flight of the light pulses, the straight-line distance between the sensor and the object is determined. Combined with the wearing geometry model of the smart glasses (e.g., the relative positional deviation between the sensor and the user's eyes), this straight-line distance is calibrated and compensated to obtain the actual viewing distance between the user's eyes and the object.

[0030] Step A20: Obtain eye image data collected by the camera sensor, and extract eye usage time based on the eye image data; The camera sensor is typically an infrared or visible light camera pointed towards the user's eyes, used to continuously capture images of the eye area. Pupil detection and blink recognition are performed on each frame of the eye image; when a pupil is detected and the eyes are open, it is determined that the user is engaged in visual activity.

[0031] The continuous screen time is obtained by accumulating the time period during which users are continuously in a state of "eye-use activity".

[0032] Step A30: Acquire motion data collected by the motion sensor and extract head posture based on the motion data; Motion sensors typically include a three-axis accelerometer and a three-axis gyroscope, which collect linear acceleration and angular velocity data of the user's head at a certain sampling frequency. By fusing the data from the accelerometer and gyroscope (e.g., using complementary filtering or extended Kalman filtering algorithms), the Euler angles of the smart glasses in three-dimensional space, such as pitch, yaw, and roll angles, are calculated. The calculated Euler angles are used as head posture features to subsequently determine whether the user has excessive head tilting, or other poor eye-use postures.

[0033] Step A40: Obtain the light intensity data collected by the ambient light sensor, and extract the ambient illuminance based on the light intensity data.

[0034] Ambient light sensors are typically mounted on the front of the frame or the outside of the temples of smart glasses to receive ambient light in the direction of the user's forward field of vision. The sensor converts the light signal into a current or voltage value, which is then converted from analog to digital to obtain a digital light intensity value, measured in lux.

[0035] In one possible implementation, the camera sensor includes an infrared camera module and an RGB camera module. The eye image data includes an infrared image of the eye captured by the infrared camera module and an RGB image of the eye captured by the RGB camera module. The step of extracting eye usage time based on the eye image data includes: Step B10: Determine the gaze direction based on the corneal reflection spot and pupil position in the infrared image of the eye; The infrared camera module emits near-infrared light towards the user's eye, forming a reflected spot (Pulchin spot) on the corneal surface, while simultaneously acquiring infrared images containing both the pupil and the spot. Image processing is performed on each frame of the infrared image to detect the center position of the pupil and the center position of the corneal reflected spot, calculating a two-dimensional offset vector between them. Using pre-calibrated user eye optical parameters (such as corneal curvature radius, the angle between the eyeball rotation center and the optical axis), this offset vector is mapped to the deflection angle of the user's gaze in the horizontal and vertical directions, thereby determining the current gaze direction.

[0036] Step B20: Based on the eye region features in the RGB image of the eye and the gaze direction, determine whether the object being gazed upon is an electronic screen; The RGB camera module captures color images including the periorbital region and extracts eye region features, such as eyelid opening and closing degree and reflective texture of the eye surface, through semantic segmentation or object detection algorithms. Simultaneously, it combines the obtained gaze direction to comprehensively determine the type of object the user is looking at. Specifically, a lightweight classification model can be pre-trained, taking the gaze direction angle and screen reflection features extracted from the eye RGB image (such as screen border reflections, periodic refresh stripes, etc.) as input, and outputting a binary classification result of "electronic screen" or "non-electronic screen".

[0037] Step B30: When the object of gaze is determined to be an electronic screen, start timing and accumulate the gaze duration on the electronic screen while the object of gaze remains an electronic screen. The classification model outputs "electronic screen," triggering a timer to start. In each subsequent frame or time slice, the object being gazed upon is continuously identified, and the timer increments as long as the object is continuously identified as an electronic screen.

[0038] Furthermore, in order to eliminate brief misjudgment jitter, an anti-shake window can be set. Only when the judgment result remains on the electronic screen for more than the duration of the window will it be officially included in the cumulative duration; brief interruptions within the window are not counted as interruptions.

[0039] Step B40: When it is determined that the object of gaze is not an electronic screen, stop the accumulation and determine the currently accumulated gaze duration on the electronic screen as the eye use duration.

[0040] If the classification model outputs "non-electronic screen," the timer pauses its accumulation but does not clear the existing accumulated value. Simultaneously, a rest-determination logic is introduced: if the duration of fixation on a non-electronic screen exceeds a preset rest threshold, it indicates that the user has actively stopped watching the electronic screen for a rest. In this case, the currently accumulated electronic screen fixation time is output as the final effective eye usage time, and the timer is reset for the next session. If the fixation time switches back to the electronic screen within the rest threshold time, the timer resumes its accumulation, achieving a seamless transition between continuous viewing times.

[0041] In one possible implementation, after the step of fusing the various eye health levels to obtain a comprehensive eye health level, the method further includes: Step C10: Collect multiple comprehensive eye health data within a continuous time window; After fusing the comprehensive eye health status at each moment (e.g., calculating a score S(t) every second), a cache queue with a fixed time window is started. Within the time window, multiple comprehensive eye health status values ​​are continuously collected and stored according to a fixed sampling period (e.g., once per second).

[0042] Step C20: Calculate the average value of the overall eye health level within the time window, and based on the average value of the overall eye health level, execute the step of outputting a prompt message if the overall eye health level is less than a preset threshold.

[0043] The arithmetic mean of all cached comprehensive eye health values ​​within the current time window is calculated to obtain the average health level within that window. This calculated average health level is then compared to a preset threshold, rather than directly using instantaneous values.

[0044] When the average eye health level falls below a preset threshold, a notification message is triggered. Using a window average effectively smooths out drastic fluctuations in eye health caused by momentary movement disturbances (such as briefly looking down or quickly scanning the screen edge), avoiding false triggers of alerts due to single outliers. It also more accurately reflects the user's overall eye health level over a continuous period, thereby improving the accuracy and reliability of alerts.

[0045] In one possible implementation, the preset threshold includes a first threshold, a second threshold, and a third threshold that decrease sequentially. The step of outputting a prompt message if the overall eye health level is less than the preset threshold includes: Step D10: If the overall eye health level is less than the first threshold and greater than or equal to the second threshold, then output a prompt message of the first intensity. The first, second, and third thresholds decrease sequentially. For example, the first threshold can be set to 85, the second threshold to 70, and the third threshold to 50. When the calculated overall eye health level S(t) is between the first and second thresholds (e.g., 70 ≤ S(t) < 85), it indicates that the user's current eye use behavior is in a mildly unhealthy state.

[0046] At this point, the first intensity of the alert is triggered, which can manifest as a gentle vibration lasting 2 seconds, accompanied by a brief flashing of the LED indicator. This first intensity alert aims to gently remind the user to pay attention to their eye usage, avoiding excessive distraction.

[0047] Step D20: If the overall eye health level is less than the second threshold and greater than or equal to the third threshold, then output a prompt message of the second intensity. When S(t) is between the second and third thresholds (e.g., 50 ≤ S(t) < 70), it indicates that the user's eye-use behavior has reached a moderately unhealthy state. At this time, a second-intensity prompt message is triggered, which can be manifested as a strong vibration lasting 5 seconds, accompanied by a brief "beep" sound.

[0048] Step D30: If the overall eye health level is less than the third threshold, then output a prompt message of the third intensity.

[0049] When S(t) is less than the third threshold (e.g., S(t) < 50), it indicates that the user's eye behavior is in a severely unhealthy state, such as prolonged close-range viewing of electronic screens in extremely poor ambient lighting. At this time, a third-intensity prompt message is triggered, which can be manifested as continuous strong vibration and prompt sound.

[0050] Furthermore, if the adverse condition persists for more than the preset mandatory lockout time (e.g., 3 minutes), the "mandatory relaxation lockout" mechanism is triggered. This includes controlling the lenses of the smart glasses to darken to force the user to rest, while simultaneously pushing the record of this serious eye-use behavior to the parent's app.

[0051] In addition, if you are in a quiet place such as a school classroom, you can turn on silent mode, which will disable all sound alerts and only retain vibration or visual alerts.

[0052] Through the above-mentioned graded intervention strategy, this embodiment only initiates mandatory intervention measures, including darkening of lenses and push notifications to parents, when the degree of eye health drops to a severe level. This avoids the interference caused by excessively frequent reminders due to minor daily bad behaviors, and can effectively block and correct when the user's eye behavior truly reaches a harmful level, thus achieving a reasonable balance between reminder comfort and prevention effectiveness.

[0053] Based on the first and / or second embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to that in embodiments one and two above can be referred to the above description, and will not be repeated hereafter. On this basis, refer to Figure 2 As shown, after the step of fusing the various eye health levels to obtain a comprehensive eye health level, the method further includes: Step E10: Calculate the vitality value of the virtual character based on the overall eye health level, wherein the vitality value is positively correlated with the overall eye health level; The real-time comprehensive eye health status (e.g., the average value S within a continuous time window) is converted into the vitality value V of the virtual avatar through a preset mapping function. This mapping function typically uses a linear positive correlation, but piecewise linear or nonlinear functions can also be used to enhance the sensitivity of low health status areas. As an example, a specific implementation method can be applied to the virtual avatar displayed on the homepage of the parent's app (users can choose from pets, plants, superheroes, etc.). In this example, the mapping rule for the vitality value is: V = base value (50) + average eye health score × 0.5.

[0054] Step E20: Determine the appearance of the virtual avatar based on the numerical range of the vitality value; The full range of vitality values ​​can be divided into multiple consecutive level intervals, and each interval is pre-configured with a set of appearance parameters, such as facial expressions, body posture, activity level, and auxiliary effects (such as gloss and color).

[0055] Based on the current vitality value's range, identify and determine the corresponding appearance. For example, vitality values ​​are divided into three ranges: when V ≥ 80, it's considered a healthy and active range; when 60 ≤ V < 80, it's considered a fatigued and sub-healthy range; and when V < 60, it's considered a sick and listless range. The appearance corresponding to each range is as follows: In the healthy and active range, the virtual character appears healthy and active (e.g., the pet's fur is shiny, the plants have lush foliage, and the superhero stands tall); in the fatigued and sub-healthy range, the character appears tired or in a sub-healthy state (e.g., the pet is listless, and the plant leaves are yellowing); and in the sick and listless range, the character appears sick or listless (e.g., the pet lies motionless, and the plant withers and droops).

[0056] Step E30: Generate or update the virtual avatar according to the determined appearance.

[0057] If the virtual avatar has not yet been created, it will be initialized and rendered based on the determined appearance parameters and then displayed. If the virtual avatar already exists, its current appearance will be updated to the new state in real time, allowing users to perceive changes in health status through continuous or step-by-step visual changes.

[0058] In a specific example of the parent-facing app, based on the current activity level range, the selected type of virtual avatar (such as a pet, plant, or superhero) is dynamically generated or updated on the app's homepage. As the user's eye health level fluctuates, the virtual avatar's appearance smoothly or abruptly switches between shiny and withered fur, and between active and lying down. This provides intuitive and fun feedback to child users on their eye health level, encouraging them to proactively adjust their eye habits to maintain the virtual avatar's "healthy" state.

[0059] In one possible implementation, after the step of generating or updating the virtual avatar according to the determined appearance, the method further includes: Step F10: If the overall eye health level is greater than or equal to the preset threshold, then accumulate eye health points; A pre-set health threshold (e.g., corresponding to the aforementioned reminder trigger threshold, S≥85) is used to determine whether a user's eye-use behavior is at an acceptable or excellent level. Whenever the real-time calculated overall eye health level (or window average S) is greater than or equal to this threshold, it is considered that the user has maintained good eye-use habits during the current time period, and therefore a certain number of eye health points are added to the user's points account.

[0060] The points accumulation rule can be a fixed increment (e.g., 1 point for each achievement) or a weighted accumulation based on the specific health level (e.g., the higher the S, the more points are added at once). These points are used to drive the growth of the virtual avatar and provide positive behavioral reinforcement incentives for users.

[0061] Step F20: When the accumulated eye health points reach a preset growth threshold, the growth stage of the virtual character is updated. The growth stages are infancy, growth, and evolution, and different growth stages correspond to different appearances of the virtual character.

[0062] Multiple consecutive growth thresholds are preset. For example, the points required to reach the evolution stage in infancy are 0-100, in growth stage it is 101-300, and in evolution stage it is 301 or more. As users accumulate eye health points, the virtual avatar advances from the current stage to the next stage when the points first cross a certain growth threshold.

[0063] Each growth stage corresponds to a unique appearance: the infancy stage is characterized by a small size, clumsy movements, and simple special effects (such as a small pet curling up or a seedling just sprouting); the growth stage is characterized by a medium size, more active movements, and richer details (such as a pet starting to run or a plant growing branches); and the evolution stage is characterized by a robust size, gorgeous special effects, or special decorations (such as a pet growing wings, plants flowering and bearing fruit, or a superhero acquiring new equipment). During updates, the virtual avatar's appearance will smoothly transition from the current stage to the new stage, possibly accompanied by celebratory animations or sound effects.

[0064] Through this points-driven, phased growth mechanism, child users can intuitively see the results of their consistent good eye habits, thus maintaining their motivation to self-manage their eye habits.

[0065] For example, in order to help understand the technical concept or technical principle of the eye behavior monitoring method after combining this embodiment with the first and second embodiments described above, a specific embodiment is listed here. In this specific embodiment, the eye behavior monitoring method is applied to an eye behavior monitoring system, which includes smart glasses and a cloud server that are connected in communication.

[0066] (a) Smart glasses terminal The smart glasses terminal includes the glasses themselves, along with a multimodal sensor array, an edge computing unit, and a data transmission module integrated on them. The edge computing unit performs preliminary processing of the sensor data and then sends the data to a cloud server via the data transmission module. The functions of each sensor are briefly described below. Ambient Light Sensor (ALS): Real-time acquisition of ambient light intensity to determine whether the illuminance of the environment for eye use meets the standard; ToF (Time-of-Flight) distance sensor: measures the distance between the eyes and the reading material / electronic screen to determine if the viewing distance is too close; Inertial Measurement Unit (IMU): Tracks head posture and determines reading / writing posture; Infrared camera: Captures fixation point, fixation distance, and blink frequency through corneal reflection and pupil changes; RGB camera: Captures the reflection of the cornea in the eye to determine whether the person is viewing an electronic screen.

[0067] (ii) Cloud server The cloud server includes a data receiving module, an eye behavior recognition module, a dynamic scoring module, and a data storage module.

[0068] Eye Behavior Recognition Module: This module receives multimodal sensor data uploaded by the smart glasses terminal and recognizes various eye behaviors, such as outdoor activity behavior, electronic product usage behavior, reading and writing posture behavior, eye distance behavior, ambient light behavior, and eye relaxation behavior.

[0069] Dynamic Scoring Module: This module generates an eye health score based on eye behavior recognition results and according to preset rules. The scoring uses a dynamic weighted algorithm, comprehensively considering factors such as behavior type, duration, and user age. Specifically, the following logic can be used: The "eye health score S(t)" is calculated using a weighted fusion algorithm:

[0070] in, It is a dynamic weighting parameter that can be adjusted based on the user's age or environment. It is a distance risk function, which is a non-linear mapping based on the ToF sensor readings. The closer the distance, the more severe the deduction. A distance greater than 40cm is a perfect score. It is a duration risk function that takes into account the judgment results of infrared cameras and RGB cameras. It adopts an integral accumulation logic. The longer the duration of close-range viewing of the electronic screen, the more linearly the score of this item decreases. It is a posture risk function that calculates head posture using IMU data, such as deducting points when the pitch angle deviates from the standard value or when there is a head tilting action. It is an illuminance risk function that triggers a deduction when the ambient light is too dim (<300 lux). The numerical range is limited to 0-100.

[0071] Dynamic sliding window mechanism: This system does not use the score at a single instant, but rather the average score over a 10-second sliding window. This is to avoid misjudgments caused by momentary data jumps due to the wearer's brief actions.

[0072] Data storage module: Stores user sensor data, behavior recognition results, historical scores, task completion records, and virtual avatar status data.

[0073] (III) The Logic of Tiered Reminders and Interventions Based on average eye health score This system triggered a four-level intervention mechanism. excellent: A score of ≥85 is considered good behavior. No reminders will be given, and the system will record the excellent behavior data and accumulate "health points". Note: 70≤ <85, judged as minor misconduct, the glasses side emits a soft vibration for 2 seconds, accompanied by a brief flashing of the LED indicator; Warning: 50≤ <70, judged as moderate bad behavior, the glasses side emits a strong vibration for 5 seconds, accompanied by a short "beep" warning sound; serious: If the value is less than 50, it is considered a serious offense. In addition to continuous strong vibration and alert sound from the glasses side, if this state continues for 3 minutes, a "forced release lock" will be triggered. If the glasses lenses darken, this record will be pushed to the parent's app.

[0074] In scenarios such as school classrooms, you can enable silent mode to disable sound alerts and only retain vibration or visual alerts.

[0075] (iv) Parent App The parent-side app may include a virtual avatar display module, an achievement system module, and a data reporting module.

[0076] Virtual Avatar Display Module: This module displays a virtual avatar (selectable types include pets, plants, superheroes, etc.) on the parent's app homepage. The virtual avatar's status is linked to the child user's eye health score in real time, with the following mapping rules for reference.

[0077] Vitality value mapping: Virtual image vitality value = base value (50) + average eye health score * 0.5.

[0078] Appearance Status Mapping: When the vitality value is ≥80, the virtual image is in a healthy and active state (e.g., the pet's fur is shiny and the plant's leaves are lush); when the vitality value is ≤60 and <80, the virtual image is in a tired or sub-healthy state (e.g., the pet is listless and the plant's leaves are yellow); when the vitality value is <60, the virtual image is in a sick or listless state (e.g., the pet is lying down and the plant is withered).

[0079] Growth Stage Mapping: When the accumulated points reach a threshold, the virtual character enters a new growth stage (such as infancy → growth stage → evolution stage), and its appearance changes significantly.

[0080] It should be noted that the virtual avatar of this invention is only displayed in the parent's app and does not issue advice or reminders to the child through smart glasses, so as to avoid interfering with the child's normal learning and life.

[0081] Achievement System Module: This module records the tasks completed and milestones achieved by the child, displaying them in the form of an achievement wall. The types of achievements that can be referenced include: Cumulative Achievements: Accumulating a certain number of points or meeting certain criteria for a certain number of consecutive days will earn you advanced achievements (such as "Eye Protection Rising Star", "Posture Model", "Outdoor Expert" etc.). Hidden Achievements: These are achievements not listed in the achievement list, but can be obtained by demonstrating exceptional performance in a particular activity (e.g., "Expert in Far-Seeing"). Achievements correspond to virtual items or outfits, which can be unlocked to decorate virtual avatars and further increase the fun.

[0082] Data Report Module: Displays children's eye-use behavior data, historical score trends, task completion status, etc. in the form of charts, making it easier for parents to understand their children's eye-use habits and provide targeted guidance.

[0083] Settings module: This allows parents to define parameters such as scoring thresholds, task goals, and virtual avatar type when using the device for the first time.

[0084] Based on the above system, referring to Figure 3 As shown, the eye use behavior monitoring process includes: First, the smart glasses terminal collects multimodal sensor data in real time. Through the ambient light sensor ALS, distance sensor ToF, inertial measurement unit IMU, infrared camera and RGB camera, it simultaneously acquires raw eye behavior data such as ambient light intensity, eye distance, head posture, gaze state and screen viewing state. Subsequently, the edge computing unit performs preliminary processing on the data and uploads it to the cloud server. The edge computing unit performs preprocessing operations such as filtering and noise reduction and time synchronization on the raw data collected by various sensors, and then uploads the processed data to the cloud server through the data transmission module. Next, the cloud server identifies the categories of eye-use behavior. The cloud-based eye-use behavior recognition module receives the uploaded multimodal sensor data and identifies various categories of eye-use behavior, such as outdoor activity behavior, electronic product usage behavior, and reading and writing posture behavior, thereby extracting eye-use behavior features. Afterwards, the dynamic scoring module generates an average eye health score. Based on the identified eye behavior categories, the dynamic scoring module calculates the eye health score through a preset dynamic weighted fusion algorithm, and combines it with a 10-second sliding window mechanism to take the average score within the window as the final average eye health score. Then, based on the average eye health score, graded intervention measures are implemented. The system triggers a four-level intervention mechanism of corresponding intensity according to the scoring range of the average eye health score. It outputs prompts such as vibration, light, and sound through smart glasses. If necessary, it triggers forced relaxation lock and pushes the record to the parent's APP. At the end of each day, the cloud server calculates the daily average score and task completion rate. The cloud server summarizes and calculates the average score of all eye health activities for the day to obtain the daily average score of eye health, and also compiles the completion status of each eye care task for the day. Finally, the parent-side app retrieves data to update the virtual avatar's vitality value and health status based on the daily average score, and unlocks achievements and virtual props based on task completion. The parent-side app retrieves the daily average score and task completion data from the cloud, updates the virtual avatar's vitality value, appearance status, and growth stage through the virtual avatar display module, unlocks corresponding achievements and virtual props through the achievement system module, and generates an eye use behavior data report through the data report module for parents to view.

[0085] It should be noted that the above examples are only used to help understand this embodiment and do not constitute a limitation on the eye behavior monitoring process of this embodiment. Any simple modifications based on this technical concept are within the protection scope of this application.

[0086] Based on the first, second, and / or third embodiments of this application, in the fourth embodiment of this application, the content that is the same as or similar to the above-described embodiments one, two, and three can be referred to the above description and will not be repeated hereafter. Based on this, the step of fusing the various eye health levels to obtain a comprehensive eye health level includes: Step G10: Obtain the weight coefficients corresponding to each of the aforementioned eye-use behavior features; An initial weighting coefficient can be pre-assigned to each visual behavior characteristic (such as viewing distance, viewing duration, head posture, and ambient light intensity) to reflect the importance of that characteristic in the comprehensive visual health assessment. The sum of all weighting coefficients is fixed at 1 (or 100%).

[0087] Step G20: Calculate the duration of each of the aforementioned eye-use behavior characteristics in an unhealthy state. Each eye-use behavior feature is monitored in real time. The current "unhealthy state" of a feature is determined by whether its health score falls below its corresponding unhealthy threshold (e.g., distance health score fD(d) below 0.4, or duration health score fT(t) below 0.5). An independent timer is started for each feature to record the duration of its continuous unhealthy state. Once the feature recovers to a healthy state (health score rises above the threshold), the corresponding timer is immediately reset; if it re-enters an unhealthy state, the timer restarts. This independent statistical approach allows for the detection of persistent adverse trends in a specific eye-use behavior dimension.

[0088] Step G30: When any of the durations exceeds the corresponding first duration threshold, the weight coefficient of the applied eye behavior feature is increased by the first step length, and the weight coefficients of other eye behavior features are proportionally reduced, so as to maintain the sum of all weight coefficients equal to a fixed total value. A corresponding first duration threshold can be set for each eye-use behavior characteristic. For example, the distance characteristic can be set to 10 seconds, the duration characteristic to 30 seconds, the posture characteristic to 15 seconds, and the illuminance characteristic to 5 seconds. When the duration of continuous unhealthy behavior of a certain characteristic (called an "abnormal characteristic") exceeds its corresponding first duration threshold, it indicates that the unhealthy behavior in that dimension has lasted for a relatively long time, and its influence weight in the comprehensive assessment needs to be increased in order to trigger reminders or take intervention measures more quickly.

[0089] Increase the weight coefficient of the anomalous feature by a preset first step size (e.g., increase by 0.1), while proportionally reducing the weight coefficients of all other features according to their current weights, so that the sum of all weight coefficients still equals a fixed total value (e.g., 1). This adjustment method ensures that increasing the weight of the anomalous feature does not disrupt the overall balance, while the relative importance of other features remains unchanged.

[0090] Step G40: The eye health status of each condition is weighted and summed according to the adjusted weighting coefficients to obtain the comprehensive eye health status.

[0091] After the dynamic adjustment of the weight coefficients is completed, the system multiplies the health score corresponding to each eye behavior feature at the current moment (e.g., fD(d), fT(t), fP(p), fL(l, with values ​​of 0~1 or 0~100)) with the adjusted weight coefficients respectively, and then sums all the product results to obtain the weighted sum of the comprehensive eye health score.

[0092] Because the weighting coefficients are adaptively adjusted according to the duration of persistent poor performance of each feature, when a certain eye behavior dimension shows persistent abnormalities, the contribution of that dimension to the overall score is amplified, causing the overall eye health level to decline more quickly, thereby shortening the reminder response time. Conversely, when each dimension experiences only short-term fluctuations, the weights remain in their initial state, avoiding drastic changes in the overall score, thus achieving more accurate and timely eye behavior monitoring.

[0093] Furthermore, this application also proposes a smart pair of glasses, referring to... Figure 4 As shown, the smart glasses include a multimodal sensor and a processor; The multimodal sensor is used to collect users' eye behavior data; The processor is used to perform the steps of the eye behavior monitoring method described above.

[0094] The smart glasses provided in this application, employing the eye behavior monitoring method described in the above embodiments, can solve the technical problem of how to improve the accuracy of eye behavior reminders and reduce invalid or missed reminders. Compared with the prior art, the beneficial effects of the smart glasses provided in this application are the same as those of the eye behavior monitoring method provided in the above embodiments, and other technical features of the smart glasses are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0095] Furthermore, to achieve the above objectives, embodiments of this application also provide an eye-use behavior monitoring system, referring to... Figure 5 As shown, the eye behavior monitoring system includes smart glasses and a cloud server connected by communication, and the smart glasses include a multimodal sensor; The multimodal sensor is used to collect users' eye behavior data; The cloud server is used to perform the steps of the eye behavior monitoring method described above.

[0096] The eye-use behavior monitoring system provided in this application, employing the eye-use behavior monitoring method described in the above embodiments, can solve the technical problem of how to improve the accuracy of eye-use behavior reminders, thereby reducing invalid or missed reminders. Compared with the prior art, the beneficial effects of the eye-use behavior monitoring system provided in this application are the same as those of the eye-use behavior monitoring method provided in the above embodiments, and other technical features of the smart glasses are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0097] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0098] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0099] In addition, to achieve the above objectives, embodiments of this application also provide a readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the eye behavior monitoring method in the above embodiments.

[0100] The computer-readable storage medium provided in this application embodiment may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0101] The aforementioned computer-readable storage medium may be included in the smart glasses; or it may exist independently and not assembled into the smart glasses.

[0102] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by smart glasses, cause the smart glasses to perform the process steps of any embodiment of the above-described eye behavior monitoring method.

[0103] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0104] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0105] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the modules themselves.

[0106] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described eye behavior monitoring method. This solves the technical problem of how to improve the accuracy of eye behavior reminders and reduce invalid or missed reminders. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the eye behavior monitoring method provided in the above embodiments, and will not be repeated here.

[0107] Furthermore, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the eye behavior monitoring method described above.

[0108] The specific implementation of the computer program product in this application is basically the same as the embodiments of the above-mentioned eye behavior monitoring method, and will not be repeated here.

[0109] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0110] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0111] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software sensor. This computer software sensor is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a smart glasses device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0112] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. An eye behavior monitoring method, characterized by, Applied to smart glasses, the smart glasses including a multimodal sensor, the eye behavior monitoring method includes the following steps: The eye-use behavior data collected by the multimodal sensor is acquired, and eye-use behavior features are extracted based on the eye-use behavior data. The eye-use behavior features include at least two of the following: eye-use distance, eye-use duration, head posture, and ambient light intensity. Assess the visual health status corresponding to each of the aforementioned visual behavior characteristics, and integrate the visual health status of each of the aforementioned visual health status to obtain a comprehensive visual health status. If the overall eye health level is less than a preset threshold, a prompt message will be output.

2. The eye use behavior monitoring method of claim 1, wherein The multimodal sensor includes a time-of-flight sensor, a motion sensor, a camera sensor, and an ambient light sensor. The step of acquiring eye-use behavior data collected by the multimodal sensor and extracting eye-use behavior features based on the eye-use behavior data includes: Acquire distance data collected by the time-of-flight sensor, and extract eye distance based on the distance data; Acquire eye image data collected by the camera sensor, and extract eye usage time based on the eye image data; The motion data collected by the motion sensor is acquired, and the head pose is extracted based on the motion data; The ambient light intensity data collected by the ambient light sensor is acquired, and the ambient light illuminance is extracted based on the ambient light intensity data.

3. The eye use behavior monitoring method according to claim 2, wherein The camera sensor includes an infrared camera module and an RGB camera module. The eye image data includes infrared images of the eye captured by the infrared camera module and RGB images of the eye captured by the RGB camera module. The step of extracting eye usage time based on the eye image data includes: The gaze direction is determined based on the corneal reflective spot and pupil position in the infrared image of the eye; Based on the eye region features in the RGB image of the eye, and in conjunction with the gaze direction, determine whether the object being gazed upon is an electronic screen; When the object of gaze is determined to be an electronic screen, the timing begins, and the gaze duration on the electronic screen is accumulated as the object of gaze remains an electronic screen. When it is determined that the object of gaze is not an electronic screen, the accumulation stops, and the currently accumulated gaze duration on the electronic screen is determined as the eye usage duration.

4. The eye behavior monitoring method of claim 1, wherein, After the step of fusing the various eye health levels to obtain a comprehensive eye health level, the method further includes: Collect multiple comprehensive eye health data within a continuous time window; Calculate the average value of the overall eye health level within the time window, and based on the average value of the overall eye health level, execute the step of outputting a prompt message if the overall eye health level is less than a preset threshold.

5. The eye behavior monitoring method of claim 1, wherein, The preset thresholds include a first threshold, a second threshold, and a third threshold that decrease sequentially. The step of outputting a prompt message if the overall eye health level is less than the preset threshold includes: If the overall eye health level is less than the first threshold and greater than or equal to the second threshold, then a prompt message of the first intensity is output; If the overall eye health level is less than the second threshold and greater than or equal to the third threshold, then a prompt message of the second intensity will be output. If the overall eye health level is less than the third threshold, a prompt message of the third intensity will be output.

6. The eye behavior monitoring method of claim 1, wherein, After the step of fusing the various eye health levels to obtain a comprehensive eye health level, the method further includes: The vitality value of the virtual character is calculated based on the overall eye health status, wherein the vitality value is positively correlated with the overall eye health status; The appearance of the virtual avatar is determined based on the numerical range of the vitality value. The virtual avatar is generated or updated according to the determined appearance.

7. The eye use behavior monitoring method of claim 6, wherein, After the step of generating or updating the virtual avatar according to the determined appearance, the method further includes: If the overall eye health level is greater than or equal to the preset threshold, then the eye health score is accumulated; When the accumulated eye health points reach a preset growth threshold, the growth stage of the virtual character is updated. The growth stages are infancy, growth, and evolution, and different growth stages correspond to different appearances of the virtual character.

8. The eye use behavior monitoring method according to any one of claims 1 to 7, characterized in that, The step of fusing the various eye health levels to obtain a comprehensive eye health level includes: Obtain the weight coefficients corresponding to each of the aforementioned eye-use behavior features; The duration of each of the aforementioned eye-use behavior characteristics continuously being in an unhealthy state was statistically analyzed. When any of the durations exceeds the corresponding first duration threshold, the weight coefficient of the applied eye behavior feature will be increased by the first step length, and the weight coefficients of other eye behavior features will be reduced proportionally to maintain the sum of all weight coefficients equal to a fixed total value. The overall eye health status is obtained by weighting and summing each of the aforementioned eye health statuses according to the adjusted weighting coefficients.

9. An intelligent eyewear, characterized in that, The smart glasses include a multimodal sensor and a processor; The multimodal sensor is used to collect users' eye behavior data; The processor is configured to perform the steps of the eye behavior monitoring method as described in any one of claims 1 to 8.

10. An eye use behavior monitoring system, the eye use behavior monitoring system comprising smart glasses and a cloud server connected in communication, the smart glasses comprising a multimodal sensor; The multimodal sensor is used to collect users' eye behavior data; The cloud server is used to perform the steps of the eye behavior monitoring method as described in any one of claims 1 to 8.

Images