Wearing state detection method, smart glasses and readable storage medium

Abstract

The application discloses a wearing state detection method, smart glasses and a readable storage medium, relates to the technical field of wearable devices, and is applied to the smart glasses. The smart glasses comprise a first type of sensor and a second type of sensor. The power consumption of the first type of sensor is lower than that of the second type of sensor. The second type of sensor at least comprises a camera sensor. The method comprises the following steps: acquiring first sensing data collected by the first type of sensor; when the first sensing data meets a preset wake-up condition, waking up the second type of sensor; acquiring second sensing data collected by the woken-up second type of sensor, wherein the second sensing data at least comprises image data collected by the camera sensor; and determining a wearing state parameter of the smart glasses based on the second sensing data, wherein the wearing state parameter comprises a wearing posture parameter and a frame deformation parameter. The application realizes synchronous detection of the wearing posture and the frame deformation under the premise of low power consumption.

Description

Technical Field

[0001] This application relates to the field of wearable device technology, and in particular to a method for detecting wearing status, smart glasses, and a readable storage medium. Background Technology

[0002] With the rapid development of smart glasses technology, smart glasses have been widely used in various scenarios, including daily wear and AR (Augmented Reality) / VR (Virtual Reality) experiences. Correct wearing posture directly affects the user's visual experience and comfort; prolonged improper wearing may also cause eye fatigue and other problems. Frame deformation affects wearing stability and the accuracy of core functions (such as eye tracking). Therefore, detecting the wearing status of smart glasses is of significant practical importance.

[0003] Currently, the wearing status detection of smart glasses mainly adopts the following methods: one is to rely on a single sensor for detection, such as using an inertial measurement unit to determine head posture or using a capacitive sensor to detect the wearing status; the other is to add a dedicated detection module, such as configuring an additional deformation sensor on the frame to achieve structural detection. Regardless of which method is used, existing detection solutions usually adopt a working mode of continuous data acquisition and real-time analysis to ensure timely detection.

[0004] However, the aforementioned continuous operating mode causes the sensor to operate under high load for extended periods, significantly increasing overall power consumption and severely compressing the battery life of smart glasses. Furthermore, existing solutions struggle to simultaneously detect both wearing posture and frame deformation, resulting in relatively limited functionality. Therefore, achieving simultaneous detection of wearing posture and frame deformation while maintaining low power consumption has become a pressing technical challenge. Summary of the Invention

[0005] The main purpose of this application is to provide a wearing status detection method, smart glasses, and a readable storage medium, aiming to solve the technical problem of how to achieve synchronous detection of wearing posture and frame deformation under the premise of low power consumption.

[0006] To achieve the above objectives, this application provides a wearing status detection method for smart glasses. The smart glasses include a first type of sensor and a second type of sensor. The power consumption of the first type of sensor is lower than that of the second type of sensor. The second type of sensor includes at least a camera sensor. The method includes the following steps: Acquire the first sensing data collected by the first type of sensor; When the first sensing data meets the preset wake-up conditions, the second type of sensor is woken up; Acquire second sensing data collected by the awakened second type of sensor, wherein the second sensing data includes at least image data collected by the camera sensor; Based on the second sensing data, the wearing status parameters of the smart glasses are determined, wherein the wearing status parameters include wearing posture parameters and frame deformation parameters. The wearing posture parameters are determined at least in part based on the position of eye feature points in the image data, and the frame deformation parameters are determined at least in part based on the frame outline shape in the image data.

[0007] In addition, to achieve the above objectives, this application also provides a smart glasses, which includes a first type of sensor, a second type of sensor and a processor, wherein the power consumption of the first type of sensor is lower than that of the second type of sensor; The first type of sensor and the second type of sensor are used to collect sensing data; The processor is used to perform the steps of the wearing status detection method as described above.

[0008] In addition, to achieve the above objectives, this application also provides a readable storage medium, which is a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the steps of the wearing state detection method as described above.

[0009] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the wearing state detection method described above.

[0010] One or more technical solutions proposed in this application have at least the following technical effects: This application's embodiments employ a hierarchical working mechanism using a low-power first-type sensor as the monitoring unit and a high-power second-type sensor as the wake-up working unit. This constructs an event-driven hierarchical wake-up mechanism. Only when the first sensing data collected by the low-power first-type sensor meets the preset wake-up conditions is the high-power second-type sensor (especially the camera sensor) triggered to start working. This breaks the high-load operation mode caused by the traditional "continuous acquisition and real-time analysis" of sensors, effectively avoiding the prolonged activation of core power-consuming devices such as the camera sensor, and reducing the average power consumption of the entire device. Simultaneously, after the second-type sensor is woken up, the positions of eye feature points and the shape of the eyeglass frame are simultaneously extracted based on the image data it collects. This allows the wearing posture parameters and eyeglass frame deformation parameters to be determined together within the same detection cycle. This eliminates the need for additional dedicated deformation sensors to avoid further increases in hardware complexity and power consumption, and achieves coordinated perception of the user's correct wearing posture and the stability of the eyeglass frame structure. Therefore, this technical solution achieves simultaneous detection of wearing posture and eyeglass frame deformation while maintaining low power consumption. Attached Figure Description

[0011] 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.

[0012] 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.

[0013] Figure 1 This is a flowchart illustrating the first embodiment of the wearing status detection method of this application; Figure 2 This is a schematic diagram of the wearing posture detection process according to an embodiment of the wearing status detection method of this application; Figure 3 This is a schematic diagram of the frame deformation detection process according to an embodiment of the wearing status detection method of this application; Figure 4 This is a schematic diagram of the eyeglass structure of the smart glasses in this application; Figure 5 This is a schematic diagram of the hardware operating environment of the device involved in the wearing status detection method in the embodiments of this application.

[0014] 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

[0015] 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.

[0016] With the rapid development of smart glasses, their integrated functions are becoming increasingly rich, and they are widely used in various scenarios such as daily wear, office assistance, and AR / VR experiences. Whether the wearing posture is correct and whether the frame is deformed directly affects the user's wearing comfort and visual experience, and may even cause problems such as eye fatigue and neck discomfort due to improper wearing for a long time. At the same time, the deformation of the frame may also cause positioning deviations of the eye-tracking camera, TOF (Time of Flight) and other sensors integrated on the glasses, thereby affecting the normal operation of core functions such as eye tracking and AR display, reducing the practicality of the product.

[0017] Currently, existing smart glasses posture detection technologies have significant limitations: most solutions rely on a single sensor, such as an IMU (Inertial Measurement Unit), which can only roughly determine head posture and cannot combine the user's eye state to achieve accurate posture recognition, making it prone to misjudgment (such as misjudging a user's active head turning as an abnormal posture); some solutions attempt to achieve posture detection by adding an additional image sensor, which not only increases hardware complexity but also significantly increases device power consumption.

[0018] In terms of frame deformation detection, existing technologies have significant shortcomings: most solutions require the addition of dedicated deformation sensors, which not only increases hardware costs but also further increases device power consumption. Moreover, the detection accuracy is limited, making it difficult to capture slight deformations (such as slight bending of the temples or slight displacement of the nose bridge). A few solutions attempt to detect deformation indirectly through a single sensor, but the detection accuracy is low, and it is easily affected by wearing posture and environmental factors, resulting in poor practicality.

[0019] In addition, existing detection methods generally suffer from excessive power consumption: to ensure real-time detection, each sensor is often in a state of continuous full-load operation, which leads to a significant decrease in the battery life of smart glasses, restricting their portability and practicality; at the same time, there is a lack of collaborative optimization among multiple sensors, resulting in data redundancy and resource waste, and the hardware resources integrated into smart glasses are not fully utilized (such as eye-tracking cameras are mostly used only for eye tracking and are not involved in wearing posture or deformation detection).

[0020] In summary, existing technologies cannot achieve "low power consumption, high precision, and low cost" simultaneous detection of wearing posture and frame deformation.

[0021] To address the aforementioned issues, the main solution of this application is as follows: acquiring first sensing data collected by a first type of sensor; waking up a second type of sensor when the first sensing data meets a preset wake-up condition; acquiring second sensing data collected by the woken second type of sensor, wherein the second sensing data includes at least image data collected by the camera sensor; and determining wearing state parameters of the smart glasses based on the second sensing data, wherein the wearing state parameters include wearing posture parameters and frame deformation parameters, the wearing posture parameters being determined at least partially based on the position of eye feature points in the image data, and the frame deformation parameters being determined at least partially based on the frame outline shape in the image data.

[0022] This application constructs an event-driven hierarchical wake-up mechanism by employing a low-power first-type sensor as the monitoring unit and a high-power second-type sensor as the wake-up working unit. Only when the first sensing data collected by the low-power first-type sensor meets the preset wake-up conditions is the high-power second-type sensor (especially the camera sensor) triggered to start working. This breaks the high-load operation mode caused by the traditional "continuous acquisition and real-time analysis" of sensors, effectively avoiding the long-term active state of core power-consuming devices such as the camera sensor, and reducing the average power consumption of the entire device. Simultaneously, after the second-type sensor is woken up, the positions of eye feature points and the shape of the eyeglass frame are simultaneously extracted based on the image data it has collected. This allows the wearing posture parameters and eyeglass frame deformation parameters to be determined together within the same detection cycle. This eliminates the need for additional dedicated deformation sensors to avoid further increases in hardware complexity and power consumption, and achieves coordinated perception of the user's correct wearing posture and the stability of the eyeglass frame structure. Therefore, this technical solution achieves simultaneous detection of wearing posture and eyeglass frame deformation while maintaining low power consumption. It should be noted that the implementing entity of the various embodiments of the wearing status detection method of this application can be a smart glasses capable of realizing the above functions, and the various embodiments of the wearing status detection method of this application do not impose specific limitations on this.

[0023] Based on this, this application proposes a wearing status detection method according to a first embodiment. In this embodiment, it is applied to smart glasses, which include a first type of sensor and a second type of sensor. The power consumption of the first type of sensor is lower than that of the second type of sensor, and the second type of sensor includes at least a camera sensor. (Refer to...) Figure 1 As shown, the wearing status detection method includes the following steps S10~S40: Step S10: Obtain the first sensing data collected by the first type of sensor; The first type of sensor remains continuously active as a low-power monitoring unit. It may include, but is not limited to, one or more combinations of ambient light sensors, capacitive wearable sensors, microswitches, or infrared proximity sensors. Through the coordinated configuration of multiple types of sensors, the sensitivity to sensing changes in the user's wearing status and behavior is improved.

[0024] The first type of sensor collects primary sensing data at a lower sampling frequency. Specifically, the ambient light sensor can collect ambient light intensity data to determine whether the current lighting conditions are sufficient to support the camera sensor in capturing images with acceptable clarity; the capacitive wear sensor can collect capacitance changes in the area where the temples contact the skin to sense the wearing contact state; the microswitch can detect the unfolded and folded state of the temples to reflect the usage status of the glasses; and the infrared proximity sensor can detect changes in the distance between the glasses and the wearer to help determine whether the wearer is in a wearing scenario.

[0025] The aforementioned first sensor data is used to initially determine whether the smart glasses are being worn, whether there is a change in posture, or whether there is user behavior that may trigger further detection needs. By using low-power sensors for continuous monitoring, the standby power consumption of the entire device can be kept at a low level while ensuring basic monitoring capabilities.

[0026] Step S20: When the first sensing data meets the preset wake-up conditions, wake up the second type of sensor; The preset wake-up conditions can be set based on the characteristic changes of the first sensor data, specifically in the form of a combination of one or more judgment logics. Corresponding threshold parameters can be configured for different types of the first type of sensor, forming a multi-dimensional wake-up judgment mechanism to avoid invalid wake-up caused by false triggering of a single sensor.

[0027] When the first sensor data meets a preset threshold condition, the wake-up mechanism is triggered. For example, when the ambient light sensor detects that the light intensity reaches a preset threshold, it indicates that the current lighting conditions are suitable for image acquisition, providing an environmental basis for waking up the camera sensor; when the capacitance value detected by the capacitive wear sensor changes from a low level to a high level and stabilizes within a preset range, it indicates that the glasses have been correctly worn on the head; when the microswitch detects that the temples have changed from a folded state to an unfolded state, it indicates that the glasses are about to be used; when the infrared proximity sensor detects that the distance to the wearer is less than a set threshold, it further confirms the effectiveness of the wearing status.

[0028] When any of the above conditions is met or a combination of preset conditions is met, a wake-up signal is generated to activate the second type of sensor.

[0029] Step S30: Obtain second sensing data collected by the second type of sensor that has been activated, wherein the second sensing data includes at least the image data collected by the camera sensor; The camera sensor can be configured to capture image data facing the user's eye area; for example, it could be an eye-tracking camera. Additionally, this second type of sensor may optionally include a depth sensor to assist in acquiring richer three-dimensional spatial information.

[0030] The second sensor data is obtained through a single acquisition or short-term continuous acquisition after wake-up, rather than long-term continuous acquisition. This limits the operating time of high-power devices while acquiring key detection information, and further controls the overall power consumption of the device.

[0031] Step S40: Based on the second sensing data, determine the wearing status parameters of the smart glasses, wherein the wearing status parameters include wearing posture parameters and frame deformation parameters, the wearing posture parameters are determined at least in part based on the position of eye feature points in the image data, and the frame deformation parameters are determined at least in part based on the frame outline shape in the image data.

[0032] The acquired image data, which is collected by a camera sensor facing the eyes, is subjected to image analysis and computation. The image data mainly covers the user's eye area and the local structure of the surrounding frame.

[0033] Pre-defined eye feature extraction algorithms can identify eye feature points in images, such as the pupil center, iris edge, inner corner of the eye, outer corner of the eye, and upper and lower eyelid edges. Based on the relative positional relationships between these feature points and their matching degree with a pre-defined standard model, wearing posture parameters are calculated.

[0034] Simultaneously, a preset contour extraction algorithm can identify local contours of the eye frame in the image, such as the inner edge of the lens rim, the bridge of the nose, and the geometric features of the area where the lens frame meets the eye. Since the camera's field of view is primarily directed towards the eye, it cannot fully capture the global contours of the bridge and temples, but frame deformation is often reflected in local geometric features, such as changes in curvature of the inner edge of the lens rim and asymmetric shifts in the bridge of the nose.

[0035] By comparing these real-time extracted local contour data with preset standard local features, the deformation deviation and deformation direction can be calculated to obtain the frame deformation parameters. These parameters are used to characterize whether the frame has twisted, bent, or loosened due to long-term use or external forces, providing data support for structural health monitoring and maintenance reminders.

[0036] By collecting image data through a single wake-up, the system simultaneously detects the wearing posture and frame deformation. This design fully utilizes the existing hardware resources of the eye-tracking camera, eliminating the need for additional dedicated deformation sensors. This reduces hardware costs and system complexity, while effectively unifying multi-parameter collaborative perception and low-power operation. Furthermore, it can output a complete wearing status assessment result within a single detection cycle.

[0037] 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. On this basis, the first type of sensor includes an ambient light sensor and a capacitance sensor, and the first sensing data includes ambient light intensity and capacitance value. After the step of acquiring the first sensing data collected by the first type of sensor, the method further includes: Step A10: If the ambient light intensity is greater than or equal to a preset light-sensing threshold and the capacitance value is not within the preset normal capacitance threshold range, then the first sensing data is determined to meet the preset wake-up condition. When the ambient light intensity is greater than or equal to the preset light-sensing threshold, it indicates that the current ambient light is sufficient to provide good lighting conditions for the camera sensor, ensuring that the subsequently acquired image data has sufficient clarity and recognizability, and avoiding image quality degradation due to insufficient light, which in turn affects the accuracy of the wearing status parameters.

[0038] Meanwhile, when the capacitance value is outside the preset normal capacitance threshold range, it indicates an abnormal contact between the temples of the glasses and the user's skin. Specifically, if the capacitance value is below the preset normal capacitance threshold range, it may indicate that the glasses are not worn correctly or are too loose; if the capacitance value is above the preset normal capacitance threshold range, it may indicate that they are worn too tightly or that the contact area is abnormal. This abnormal state indicates a potential possibility of frame deformation or improper wearing posture, requiring further testing and confirmation.

[0039] When both conditions are met simultaneously—ambient light intensity greater than or equal to a preset light-sensing threshold and the capacitance value not falling within a preset normal capacitance threshold range—the first sensing data is determined to satisfy the preset wake-up condition. This dual-condition joint judgment mechanism ensures that the high-power second-type sensor is only activated when both a good image acquisition environment and actual detection needs exist, thereby further optimizing overall power consumption while ensuring the necessity of detection.

[0040] Step A20: If the ambient light intensity is less than the preset light-sensing threshold, or the capacitance value is within the preset normal capacitance threshold range, then it is determined that the first sensing data does not meet the preset wake-up condition.

[0041] When the ambient light intensity is less than the preset light sensing threshold, it indicates that the current ambient light is insufficient. Even if the camera sensor is woken up, it will be difficult to obtain clear and effective image data, which cannot support the accurate determination of subsequent wearing posture parameters and frame deformation parameters. At this time, waking up the second type of sensor will cause unnecessary energy consumption.

[0042] When the capacitance value is within the preset normal capacitance threshold range, it indicates that the temples are in good contact with the skin, the glasses are in a stable wearing state, and there is no contact state deviation caused by abnormal tightness or frame deformation. At this time, there is no need for further testing.

[0043] If either of the above conditions is met—that is, the ambient light intensity is less than a preset light-sensing threshold, or the capacitance value is within a preset normal capacitance threshold range—and the first sensing data is determined not to meet the preset wake-up condition, the second type of sensor remains in sleep mode. This condition exclusion mechanism enables low-power operation in numerous scenarios where detection is unnecessary, effectively avoiding redundant wake-ups caused by unsuitable environments or proper wear.

[0044] In one possible implementation, the second type of sensor further includes a motion sensor and a time-of-flight sensor, and the step of waking up the second type of sensor includes: Step B10: When the first sensing data meets the preset wake-up condition, wake up the motion sensor; After the initial sensor data meets the preset wake-up conditions, the motion sensor is woken up first, rather than directly waking up the camera sensor, which has the highest power consumption. The motion sensor may include an inertial measurement unit or a combination of an accelerometer and a gyroscope. Its power consumption is lower than that of the camera sensor. As an intermediate wake-up node, it plays a transitional role between the preceding and following layers.

[0045] Once awakened, the motion sensor enters its working state, completing data acquisition with a shorter sampling time to avoid the additional power consumption caused by prolonged operation. This tiered wake-up design allows for secondary judgment by leveraging the low-to-medium power consumption characteristics of the motion sensor after the initial low-power sensor is triggered, rather than directly activating the high-power camera sensor. This achieves more refined energy management while ensuring detection accuracy.

[0046] Step B20: Acquire motion data collected by the motion sensor, determine the current head posture based on the motion data, and wake up the camera sensor and the time-of-flight sensor when the deviation of the current head posture from the preset reference head posture is greater than or equal to the preset posture threshold.

[0047] Motion data collected by motion sensors can be acquired, including triaxial acceleration and triaxial angular velocity data. By performing attitude calculations on the above data, the current head posture of the smart glasses can be determined, such as attitude parameters like pitch, yaw, and roll angles relative to the direction of gravity.

[0048] The determined current head posture is compared with a preset baseline head posture to calculate the deviation. The preset baseline head posture can be pre-calibrated based on the user's standard posture during normal wear, or it can be dynamically updated based on historical wearing data. When the deviation is greater than or equal to the preset posture threshold, it indicates that there is a significant abnormality in the current head posture, such as the glasses being significantly tilted, slipped, or misaligned. In this case, the probability of frame deformation or improper wearing posture is high, and further visual inspection is required.

[0049] Under these conditions, the camera sensor and time-of-flight sensor are activated. The camera sensor is used to collect image data of the eye and frame area to support the calculation of wearing posture parameters and frame deformation parameters; the time-of-flight sensor is used to acquire depth information, which can help measure the spatial distance distribution between the frame and the face, providing supplementary data support for the three-dimensional quantification of frame deformation.

[0050] If the deviation of the current head posture from the preset baseline head posture is less than the preset posture threshold, it indicates that the head posture is basically normal and there is no need for further detection. In this case, the camera sensor and time-of-flight sensor do not need to be woken up, and the motion sensor can return to sleep mode after completing this judgment. Through this secondary screening mechanism based on head posture deviation, the activation of high-power sensors can be avoided in many scenarios where the head posture is normal, further reducing the overall power consumption of the device and realizing a refined hierarchical wake-up strategy from low-power sensors to medium-power sensors and then to high-power sensors.

[0051] 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, the step of determining the wearing status parameters of the smart glasses based on the second sensor data includes: Step C10: When the wearing status parameters include wearing posture parameters, acquire the image data collected by the camera sensor, the motion data collected by the motion sensor, and the distance data collected by the time-of-flight sensor. The distance data includes the distance data in front of the temple and the distance data to the side of the temple. The distance data in front of the temple is collected by the time-of-flight sensor located at the crossbeam of the frame, and the distance data to the side of the temple is collected by the time-of-flight sensor located at the temple. When it is necessary to determine the wearing posture parameters, multi-data information collected by three types of sensors is acquired to form a multi-dimensional perception of the wearing status. Specifically, image data collected by the camera sensor is used to provide visual information about the eye area, motion data collected by the motion sensor is used to reflect the overall head posture, and distance data collected by the time-of-flight sensor is used to provide the spatial distance distribution between the frame and the face.

[0052] The time-of-flight sensors can be deployed in a zoned configuration: the time-of-flight sensor located on the frame beam collects data on the distance in front of the temples, mainly used to measure the distance between the bridge of the nose and the root of the nose and the face. This data can reflect the vertical wearing height of the glasses and the balance between the left and right sides; the time-of-flight sensor located on the temples collects data on the distance to the sides of the temples, mainly used to measure the distance between the inner side of the temples and the skin of the temporal region. This data can reflect the tightness of the temples and whether the pressure distribution on the left and right sides is symmetrical.

[0053] Through the aforementioned multi-location deployment, the time-of-flight sensor can achieve three-dimensional perception of the wearing status from a spatial geometric dimension.

[0054] Step C20: Extract eye feature data based on the image data, and extract head pose data based on the motion data; Feature extraction processing is performed on the collected multi-data. Eye feature data is extracted based on image data, which may include the pupil center position, used to determine the direction of gaze and the focal point of both eyes; iris position, used to help determine the position of the eyeball relative to the eye socket; and interpupillary distance, i.e., the distance between the centers of the pupils of both eyes, which reflects whether wearing glasses causes the line of sight of both eyes to be aligned with the optical center of the lenses.

[0055] Head posture data is extracted based on motion data. Specifically, this includes head tilt angle, i.e., the degree of left-right tilt of the head relative to the direction of gravity, used to determine whether the glasses are not worn horizontally due to abnormal head posture; and head rotation angle, i.e., the degree of pitch and yaw of the head relative to the torso, used to distinguish whether the wearing deviation is caused by head posture or structural deformation of the glasses themselves. The above head posture data can effectively exclude changes in wearing status caused by the user's own head movements, avoiding misjudgment as frame deformation or improper wearing.

[0056] Step C30: Based on the eye feature data, the head posture data, and the distance data, generate the wearing posture parameters.

[0057] The wearing posture parameters are generated by fusing eye feature data, head posture data, and distance data. In one exemplary implementation, the horizontal alignment and vertical offset of the frame relative to the line of sight of both eyes are calculated based on the pupil center position and interpupillary distance data, combined with the tilt angle in the head posture data for compensation and correction. Based on the iris position and the distance to the front of the temple, it is determined whether the frame is in a symmetrical wearing state, and whether there is unilateral forward or backward tilting. Based on the lateral distance data of the temple, combined with the interpupillary distance change trend in the eye feature data, the symmetry of the temple clamping force is evaluated, and it is determined whether there is unilateral excessive tightness or looseness due to frame deformation.

[0058] Through collaborative analysis and cross-validation of the aforementioned multi-source data, wearing posture parameters can be comprehensively derived. These parameters may include detailed indicators such as frame horizontality deviation, lens-to-line of vision verticality deviation, left-right wearing height difference, and temple clamping symmetry. Compared to analysis methods based on a single data source, this multi-sensor fusion method effectively improves the accuracy and robustness of determining wearing posture parameters, avoiding misjudgments caused by limited observation angles of a single sensor or environmental interference, thereby providing users with more reliable wearing status assessment results.

[0059] In one possible implementation, the wearing posture parameter includes a wearing posture standard, and the step of generating the wearing posture parameter based on the eye feature data, the head posture data, and the distance data includes: Step D10: Calculate the first deviation between the eye feature data and the preset benchmark eye feature, the second deviation between the head posture data and the preset benchmark head posture, and the third deviation between the distance data and the preset benchmark distance. The deviation between the three types of data and their respective preset benchmarks is calculated. Specifically, the first deviation between the eye feature data and the preset benchmark eye features is calculated. The preset benchmark eye features can be calibrated based on parameters such as the pupil center position, iris position, and interpupillary distance under standard posture in the user's historical wearing data, or set according to ergonomic standards at the factory. The first deviation can be calculated using metrics such as Euclidean distance or Mahalanobis distance to comprehensively measure the overall offset of eye features in two-dimensional image space, such as the horizontal angle deviation of the line connecting the pupil centers relative to the preset benchmark, and the scaling ratio deviation of the interpupillary distance relative to the standard value.

[0060] The second deviation between the head posture data and the preset baseline head posture is calculated. The preset baseline head posture is typically a standard posture with the head upright and looking straight ahead, which can be represented by the three-axis angle values ​​of pitch, yaw, and roll. The second deviation is obtained by weighted summation of the angular deviations along each axis, where the roll deviation reflects the horizontal tilt of the glasses, the pitch deviation reflects the vertical tilt of the glasses, and the yaw deviation reflects the left and right turning of the head. The weights of different axes can be configured according to the actual sensitivity of the wearing posture detection.

[0061] The third deviation is calculated between the distance data and the preset reference distance. The distance data includes the frontal distance of the temples collected by the time-of-flight sensor at the frame beam position, and the lateral distance of the temples collected by the time-of-flight sensor at the temple position. The preset reference distance is the standard distance value between each measuring point and the face when the user wears the glasses normally, and can be set based on the self-learning calibration during the user's first wearing or general ergonomic data. The calculation of the third deviation comprehensively considers the left-right symmetry deviation of the frontal distance of the temples and the left-right symmetry deviation of the lateral distance of the temples, such as the weighted combination of the difference between the frontal distance of the left and right temples and the difference between the lateral distances of the left and right temples. This deviation directly reflects the symmetry of the frame wearing and the balance of the clamping force.

[0062] Step D20: Combine the first deviation, the second deviation, and the third deviation to obtain the overall posture deviation. The fusion method can employ a weighted fusion strategy, assigning corresponding weights based on the confidence levels of different data sources in the wearing posture assessment. For example, under sufficient lighting and clear image conditions, eye feature data has high accuracy and can be assigned a higher weight to the first bias; under conditions of vigorous head movement, the data from the motion sensor may experience instantaneous fluctuations, and the weight of the second bias can be reduced accordingly; the distance data from the time-of-flight sensor is less affected by ambient lighting and can maintain a stable weight contribution in various scenarios.

[0063] The fusion calculation can be implemented using a weighted summation method, i.e., overall posture deviation = α × first deviation + β × second deviation + γ × third deviation, where α, β, and γ are preset weight coefficients and satisfy α + β + γ = 1. Through the above fusion mechanism, visual information, posture information, and spatial distance information can be integrated to form a quantitative evaluation of the overall deviation of the wearing posture.

[0064] Step D30: Obtain the wearing posture standard degree based on the posture comprehensive deviation degree, wherein the wearing posture standard degree is negatively correlated with the posture comprehensive deviation degree.

[0065] Wearing posture standard is obtained based on the overall posture deviation, where the wearing posture standard is negatively correlated with the overall posture deviation. A linear mapping can be used for the conversion; for example, if the maximum allowable value for the overall posture deviation is set to D_max, then the wearing posture standard = (1 - overall posture deviation / D_max) × 100%. When the overall posture deviation is 0, the standard is 100%, and when it reaches or exceeds the maximum allowable value, the standard is 0%. Alternatively, a non-linear mapping, such as an exponential decay function or a piecewise function, can be used to ensure that the standard decreases more gradually when the deviation is small, and decreases rapidly when the deviation approaches a critical value, thus better reflecting the user's actual perception of wearing comfort.

[0066] The generated standard of wearing posture can be presented to the user as an intuitive output indicator, indicating the quality of the current wearing posture. When the standard is lower than a preset threshold, a wearing adjustment prompt can be triggered to guide the user to correct their posture. At the same time, this standard can also serve as a reference for subsequent frame deformation detection, distinguishing between improper wearing posture and abnormal conditions caused by frame structural deformation, thereby improving the accuracy and interpretability of the detection results.

[0067] In one possible implementation, refer to Figure 3 As shown, the step of determining the wearing status parameters of the smart glasses based on the second sensor data includes: Step E10: When the wearing status parameters include the frame deformation parameters, acquire the image data collected by the camera sensor, the distance data collected by the time-of-flight sensor, and the capacitance distribution data collected by the capacitance sensor, wherein the capacitance distribution data is a dataset formed by arranging the capacitance values ​​output by multiple capacitance sensors deployed at different locations in a preset spatial order. When it is necessary to determine the deformation parameters of the eyeglass frame, image data acquired by a camera sensor, distance data acquired by a time-of-flight sensor, and capacitance distribution data acquired by a capacitance sensor are obtained. Specifically, the camera sensor is an eye-tracking camera facing the eyes, and the image data it acquires is used to provide contour data of the eye region and the eyeglass frame; the distance data from the time-of-flight sensor is used to provide the spatial distance distribution between the eyeglass frame and the face; and the capacitance sensor is used to provide pressure distribution information in the area where the eyeglass frame contacts the skin.

[0068] Capacitance distribution data is a dataset formed by arranging the capacitance values ​​output by multiple capacitance sensors deployed at different locations in a preset spatial order. Specifically, capacitance sensors can be deployed at multiple key contact locations such as the inner side of the temples and the nose pad area. The capacitance values ​​output by each sensor are arranged in order of their physical position on the frame, forming serialized data that reflects the spatial distribution of contact pressure between the frame and the face. When the frame deforms, the contact pressure distribution at different locations will change accordingly, and the capacitance distribution data can sensitively capture these changes.

[0069] Step E20: Extract frame outline data from the image data and extract temple lateral distance data from the distance data, wherein the temple lateral distance data is collected by a time-of-flight sensor deployed at the temple position; Feature extraction processing was performed on the collected multi-source data. Lens frame contour data was extracted from the image data acquired by the eye-tracking camera. It should be noted that since the eye-tracking camera's field of view is primarily directed towards the eyes, the complete lens outline may not be captured during actual acquisition. However, the contour features of the inner edge of the lens frame adjacent to the eyes, the area connecting to the bridge of the nose, and the lower edge of the lens frame near the eyelid can be extracted as the main components of the lens frame contour data.

[0070] Lateral temple distance data is extracted from distance data collected by time-of-flight sensors. This data, collected by time-of-flight sensors located on the temples, specifically reflects the gap between the inner side of the temple and the skin of the temporal region. Additionally, forward temple distance data can be extracted from a time-of-flight sensor located on the frame bridge, reflecting the distance between the bridge of the nose and the face. Under normal wearing conditions, the lateral and forward temple distances on both sides should be approximately symmetrical, and the values ​​should remain stable within a preset range. When the frame deforms, such as temples widening or narrowing, inconsistent opening angles of the left and right temples, or drooping of the bridge of the nose, the above distance data will exhibit abnormal increases or decreases on one side, or asymmetry between the left and right sides, which can be used as a basis for judging frame deformation.

[0071] Step E30: Based on the frame contour data, the temple lateral distance data, and the capacitance distribution data, generate the frame deformation parameters.

[0072] Based on a comprehensive analysis of frame contour data, temple lateral distance data, and capacitance distribution data, frame deformation parameters are generated. In one exemplary implementation, geometric analysis can be performed on the frame contour data. Taking the field of view of the eye-tracking camera as an example, indicators such as curvature anomalies on the inner edge of the lens rim, asymmetric offsets of the left and right contours, and midline offsets of the nose bridge contour segment can be identified to preliminarily determine whether there are signs of deformation such as frame distortion, nose bridge sagging, or local bending. It should be understood that the specific content of the frame contour data may vary depending on the actual field of view of the camera sensor; the core is to extract observable geometric features of the frame for deformation analysis.

[0073] Subsequently, the data on the lateral distance of the temples was used for verification and expansion. If the data on the lateral distance of the temples showed that the difference in distance between the left and right sides exceeded the preset symmetry threshold, or if the distance on one side was significantly offset from the baseline value, it would further confirm that the frame was deformed and could pinpoint the location of the deformation (such as the left temple widening outwards and the right temple narrowing inwards).

[0074] Finally, capacitance distribution data was introduced for cross-validation. When the frame deforms, the contact pressure distribution between the inner temple and the face changes, and the capacitance value at the corresponding location in the capacitance distribution data will fluctuate abnormally. For example, if the left temple expands outward, the capacitance value output by the capacitance sensor on that side may decrease significantly; if the nose pad area deforms, the output of the capacitance sensor at the nose pad location may show asymmetrical changes. By performing spatiotemporal alignment and fusion analysis of capacitance distribution data with frame contour data and distance data, the system can comprehensively determine frame deformation from three dimensions: geometric shape, spatial distance, and contact pressure.

[0075] The results of the above multi-source fusion analysis are ultimately quantified into frame deformation parameters, which may include deformation type (such as twisting deformation, bending deformation, outward expansion deformation, inward expansion deformation, nose bridge depression, etc.), deformation location (such as left temple, right temple, nose bridge area, local area of ​​lens ring, etc.), and deformation degree (such as mild deformation, moderate deformation, severe deformation).

[0076] In one possible implementation, the frame deformation parameters include the degree of deformation, and the step of generating the frame deformation parameters based on the frame contour data, the temple lateral distance data, and the capacitance distribution data includes: Step F10: Calculate the fourth deviation between the frame outline data and the preset reference frame outline, the fifth deviation between the temple lateral distance data and the preset reference temple lateral distance, and the sixth deviation between the capacitance distribution data and the preset reference capacitance distribution data. The fourth deviation between the frame contour data and the preset reference frame contour is calculated. The preset reference frame contour can be generated based on the standard geometric model of the smart glasses at the factory, or it can be personalized by combining the contour features in a normal state collected by the eye-tracking camera when the user first wears the glasses.

[0077] Considering that only a portion of the frame's outline may be captured during actual acquisition, the calculation of the fourth deviation can focus on the observable outline area within the field of view. This includes factors such as the curvature change of the inner edge of the lens, the symmetry deviation of the visible outlines on both sides, the central axis offset of the nasal bridge outline segment, and the shape deviation of the local arc along the lower edge of the frame. These sub-deviations can be weighted and summed or calculated using vector distance to obtain a comprehensive fourth deviation, which quantifies the degree of deviation of the observable parts of the frame from the standard shape.

[0078] The fifth deviation is calculated between the temple lateral distance data and a preset baseline temple lateral distance. The preset baseline temple lateral distance is the standard gap value between the inner sides of the left and right temples and the temporal skin when the user wears the glasses normally. It can be based on self-learning calibration during the initial wearing period or set according to ergonomic statistics. The calculation of the fifth deviation mainly considers the absolute and relative deviations of the distances on both sides. Specifically, it can include the deviation of the left temple lateral distance relative to the left baseline value, the deviation of the right temple lateral distance relative to the right baseline value, and the comparison of the difference between the left and right distances with a preset symmetry threshold. When the frame undergoes outward expansion, inward contraction, or twisting deformation, the temple lateral distance will change accordingly, and the fifth deviation can effectively capture such structural anomalies.

[0079] The sixth deviation is calculated between the capacitance distribution data and the preset baseline capacitance distribution data. The capacitance distribution data is formed by arranging the capacitance values ​​output by multiple capacitance sensors located at different positions (such as the inner temple and nose pad area) in a preset spatial order, constituting a high-dimensional feature vector reflecting the pressure distribution between the frame and the face. The preset baseline capacitance distribution data consists of the standard output values ​​of each capacitance sensor when the user is wearing the glasses normally, and can be obtained based on self-learning data collected during initial wear. The sixth deviation can be calculated using a vector distance metric, comprehensively considering the capacitance value deviations at each sensor position, and assigning higher weights to key contact areas (such as the nose pad and the middle section of the temple). When the frame deforms, the contact pressure distribution at different positions will change accordingly, and the sixth deviation can sensitively reflect the overall shift and local anomalies in the contact pressure distribution.

[0080] Step F20: Combine the fourth deviation, the fifth deviation, and the sixth deviation to obtain the overall shape deviation. The fourth, fifth, and sixth deviation values ​​are fused to obtain the overall shape deviation value. A weighted fusion strategy can be used, assigning appropriate weights based on the sensitivity and reliability of different data sources in frame deformation detection. For example, frame contour data directly reflects the overall geometry of the lens rim and has high sensitivity to torsional and bending deformations, thus the fourth deviation value can be given a higher weight. Temple lateral distance data directly responds to deformation types such as temple expansion and contraction, thus the fifth deviation value can be given a moderate weight. Capacitance distribution data reflects the impact of deformation on wearing comfort from the perspective of contact pressure distribution and is less affected by ambient light, thus the sixth deviation value can be given an appropriate weight, especially as an important supplementary basis when visual detection conditions are poor.

[0081] The specific implementation of the fusion calculation can adopt a weighted summation method, that is, the shape comprehensive deviation degree = δ × fourth deviation degree + ε × fifth deviation degree + ζ × sixth deviation degree, where δ, ε, and ζ are preset weight coefficients and satisfy δ + ε + ζ = 1.

[0082] Step F30: Determine the degree of deformation based on the overall shape deviation, wherein the degree of deformation is positively correlated with the overall shape deviation.

[0083] The degree of deformation of the frame is determined based on the overall shape deviation, where the degree of deformation is positively correlated with the overall shape deviation. The specific conversion method can employ piecewise linear or nonlinear mapping, dividing the range of the overall shape deviation into multiple intervals, each corresponding to a deformation level. For example, mild deformation, moderate deformation, and severe deformation intervals can be set. When the overall shape deviation is below a first deformation threshold, it is considered mild deformation; between the first and second deformation thresholds, it is considered moderate deformation; and above the second deformation threshold, it is considered severe deformation.

[0084] The determination of the degree of deformation can be further refined based on the type of deformation. For example, for rim distortion, which significantly impacts eye-tracking accuracy, a higher deformation level can be assigned under the same overall shape deviation. For slight temple flare deformation, if it does not affect wearing stability and can be corrected, the deformation level rating can be appropriately lowered. Corresponding maintenance reminders or adaptive adjustment commands can be generated based on the degree of deformation. When the deformation reaches a moderate or higher level, a user prompt is triggered, suggesting professional maintenance or replacement of parts. When the deformation is in the mild range, the deformation trend can be recorded for subsequent health monitoring reference.

[0085] Through the aforementioned multi-source deviation calculation, fusion, and mapping mechanism, information from three dimensions—frame contour, spatial distance, and contact pressure—can be comprehensively quantified into a unified deformation degree index. Compared to detection methods based on a single data source, this multi-dimensional fusion method significantly improves the accuracy and robustness of deformation detection, effectively distinguishing different types of deformation and quantifying their severity, thus providing reliable data support for structural health monitoring and maintenance decisions for smart glasses.

[0086] In one possible implementation, prior to the step of acquiring the first sensing data collected by the first type of sensor, the method further includes... Step G10: Collect baseline data under normal wearing conditions; When a user uses a sensor for the first time or executes a calibration command, the parameters of each sensor can be calibrated to eliminate the sensor's own zero-point drift, environmental interference, and individual differences. For example: For ambient light sensors, the current output is collected as the dark current zero-point offset value under dark room conditions where the photosensitive window is blocked, and the output is collected under a known standard light source to establish a linear mapping relationship between light intensity and electrical signal, thereby completing gain calibration.

[0087] For capacitive sensors, the capacitance values ​​of each channel are collected as baseline offset when the smart glasses are not in contact with any object (no-load state), and the range is calibrated under a standard load simulating human contact to ensure the accurate conversion relationship between capacitance value and contact pressure.

[0088] For motion sensors (inertial measurement units), the smart glasses are placed on a horizontal platform to collect the outputs of the accelerometer and gyroscope during a long period of static state. The bias and scaling factor errors of each axis are calculated, and a six-sided calibration method is used to eliminate inter-axis coupling errors to obtain calibrated attitude calculation parameters.

[0089] For the camera sensor (eye-tracking camera), automatic focus calibration is performed to determine the optimal focal length, white balance calibration is performed to adapt to different color temperature environments, and distortion correction is performed by acquiring images from a standard calibration board to obtain the lens distortion coefficient and intrinsic parameter matrix, ensuring the accuracy of subsequent image measurements.

[0090] For time-of-flight sensors, output values ​​are collected at unobstructed standard distances (such as 10cm, 20cm, and 30cm), compared with known distances, and a distance measurement error compensation model is established to complete offset calibration and nonlinear correction.

[0091] After completing the parameter calibration of the above sensors, the baseline data acquisition continues. The user is prompted to wear the smart glasses in a standard posture, keeping the head upright, eyes looking straight ahead, and the glasses horizontal and symmetrically fitted to the face. In this state, each sensor initiates a brief data acquisition: the ambient light sensor collects the current ambient light intensity as a reference for subsequent lighting condition assessment; the capacitance sensor collects multi-point capacitance values ​​of the contact areas such as the inner temples and nose pads to form a capacitance distribution baseline under normal wearing conditions; the motion sensor collects triaxial acceleration and angular velocity data under stationary conditions, and the calibrated attitude calculation yields the baseline attitude angle of the head relative to the direction of gravity; the eye-tracking camera collects images including eye features and the frame outline, and the calibrated distortion correction parameters are used to extract contour features such as the pupil center, iris position, interpupillary distance, and the inner edge of the frame and the bridge of the nose; the time-of-flight sensor collects data on the distance in front of the temples at the frame beam position and the distance to the sides of the temples at the temple position, and the calibrated distance compensation model is used to obtain the standard spatial distance between each measuring point and the face under normal wearing conditions. The above-mentioned benchmark data can be collected repeatedly and the average value is taken to eliminate random errors.

[0092] Step G20: Based on the reference data, set the light threshold corresponding to the ambient light sensor, the capacitance threshold range and reference capacitance distribution data corresponding to the capacitance sensor, the reference head posture corresponding to the motion sensor, the reference eye features and reference frame outline corresponding to the camera sensor, and the reference distance corresponding to the time-of-flight sensor, wherein the reference distance includes the reference frame front distance and the reference temple side distance. Based on the collected baseline data, the detection threshold and baseline parameters for each sensor are set respectively.

[0093] For ambient light sensors, a preset light-sensing threshold is set based on the ambient light intensity value in the reference data. A certain amount of redundancy can be added on this basis to ensure that the reliability of detection can still be maintained when the light intensity is slightly lower than the reference value.

[0094] For capacitive sensors, a preset normal capacitance threshold range is set based on the output values ​​of each capacitive sensor under normal wearing conditions. The upper and lower thresholds are usually determined by adding or subtracting the allowable deviation from the reference value. At the same time, the capacitance distribution data obtained from the reference acquisition is saved as preset reference capacitance distribution data for comparison and reference during subsequent deformation detection.

[0095] For motion sensors, the attitude angles calculated from the reference data are saved as preset reference head attitudes, such as pitch angle, yaw angle and roll angle reference values, as reference zero points to determine whether the head attitude has deviated significantly.

[0096] For camera sensors, preset benchmark eye features (such as standard pupil center position, iris position, standard interpupillary distance, etc.) and preset benchmark frame contours (such as standard frame inner edge curve, bridge of nose area contour, etc.) are extracted and saved from benchmark image data.

[0097] For the time-of-flight sensor, the distances to the front of the frame beam and to the side of the temples collected from the reference data are saved as preset reference distances to the front of the frame and to the side of the temples, respectively. This completes the parameter calibration and reference data acquisition for each sensor.

[0098] The above threshold and benchmark parameters can be customized to suit the differences in facial structure and wearing habits of different users, thereby improving the accuracy of subsequent detection.

[0099] Step G30: After calibration, control the motion sensor, the camera sensor, and the time-of-flight sensor to enter a sleep state.

[0100] After completing parameter calibration and baseline data acquisition for all sensors, sleep control is implemented for high-power sensors. Meanwhile, the first type of sensors (ambient light sensor, capacitive sensor, etc.) remain in a low-power active state, continuously monitoring the wearing status and environmental changes, providing basic monitoring capabilities for subsequent event-driven wake-up.

[0101] For example, to help understand the technical concept or principle of the wearing state detection method after combining this embodiment with the first and second embodiments described above, a specific embodiment is now listed. In this specific embodiment, the wearing state detection process includes: Initialization and calibration: When the user puts on the glasses and powers on the system, the system starts initialization and completes the baseline data acquisition and threshold setting of the eye-tracking camera, CAP, TOF, IMU, and light sensor respectively (such as light sensor wake-up threshold of 50 lux, CAP capacitance threshold of baseline value of 30±10pF, and IMU attitude threshold of ±5°). After calibration, the high-power sensors go into sleep mode, and only the light sensor and CAP sensor are monitored for low power consumption.

[0102] ② Wake-up trigger: The light sensor detects an ambient light intensity of 80 lux (≥50 lux), and the CAP sensor detects a capacitance value of 45 pF on the inner side of the temple (<40 pF, the lower limit of the baseline threshold), triggering the first level of wake-up, which wakes up the IMU sensor; the IMU collects data and calculates the head tilt angle of 7° (≥5°), triggering the second level of wake-up, which wakes up the eye-tracking camera and the TOF sensor.

[0103] ③ Wearing posture detection: The eye-tracking camera acquires eye images and extracts the pupil center offset of 0.3mm and iris offset of 0.2mm, indicating abnormal eye posture; the IMU detects a 7° forward head tilt, excluding active head turning (no sudden change in angular velocity), confirming abnormal head posture; the TOF detects a 0.6mm deviation between the frame and the center of the eyebrows and a 0.3mm difference between the temples and the cheekbones on both sides, indicating asymmetrical looseness in the fit; after data fusion, the wearing posture is determined to be abnormal (forward head tilt + asymmetrical looseness in the fit).

[0104] ④ Frame deformation detection: The eye-tracking camera extracts the frame outline, and the deviation from the baseline is 0.2mm. The temple spacing changes by 0.3mm, indicating slight temple deformation. The TOF detection shows a temple length change of 0.2mm (<threshold). Combined with the CAP sensor capacitance distribution deviation of 10pF (≥threshold), the slight frame deformation is confirmed.

[0105] ⑤ Result feedback and power consumption optimization: The processing unit controls the glasses to vibrate twice, and the voice prompts "Tilting the glasses forward, the temples are slightly deformed, please adjust your wearing posture"; after the detection is completed, the high-power sensor goes into sleep mode, while the light sensor and CAP sensor continue to monitor. Since the user is stationary, the monitoring interval is adjusted from 100ms to 500ms to reduce power consumption.

[0106] ⑥ Loop detection: After the user adjusts the wearing posture, the CAP sensor detects that the capacitance value has returned to normal. When the IMU wakes up again, it detects a head posture deviation of 3° (< threshold). The second level of wake-up is not triggered, and the system goes into sleep mode after working for 1 second. The system continues to monitor in a loop until an abnormality is detected again or the user takes off the glasses.

[0107] It should be noted that the above examples are only used to help understand this embodiment and do not constitute a limitation on the wearing status detection method of this embodiment. Any simple modifications based on this technical concept are within the protection scope of this application.

[0108] Furthermore, embodiments of this application also propose a smart pair of glasses, referring to... Figure 4 As shown, the smart glasses include a first type of sensor, a second type of sensor, and a processor, wherein the power consumption of the first type of sensor is lower than that of the second type of sensor; The first type of sensor and the second type of sensor are used to collect sensing data; The processor is used to perform the steps of the wearing status detection method as described above.

[0109] like Figure 5As shown, the edge device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the edge device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the edge device to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows edge devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0110] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0111] The edge device provided in this application, employing the wearing state detection method described in the above embodiments, can solve the technical problem of how to simultaneously detect wearing posture and frame deformation under low power consumption. Compared with the prior art, the beneficial effects of the edge device provided in this application are the same as those of the wearing state detection method provided in the above embodiments, and other technical features of this edge device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0112] 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.

[0113] 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.

[0114] In addition, to achieve the above objectives, this application also provides a readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the wearing state detection method in the above embodiments.

[0115] 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.

[0116] The aforementioned computer-readable storage medium may be included in the edge device; or it may exist independently and not assembled into the edge device.

[0117] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by an edge device, cause the edge device to perform the process steps of any embodiment of the aforementioned wearing state detection method.

[0118] 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).

[0119] 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.

[0120] 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.

[0121] 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 wearing state detection method. This solves the technical problem of how to achieve simultaneous detection of wearing posture and frame deformation under low power consumption. 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 wearing state detection method provided in the above embodiments, and will not be repeated here.

[0122] Furthermore, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the wearing state detection method as described above.

[0123] The specific implementation of the computer program product in this application is basically the same as the embodiments of the above-described wearing status detection method, and will not be repeated here.

[0124] 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.

[0125] 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.

[0126] 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 an edge 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.

[0127] 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. A method for detecting wearing status, characterized in that, The method is applied to smart glasses, which include a first type of sensor and a second type of sensor. The first type of sensor has lower power consumption than the second type of sensor, and the second type of sensor includes at least a camera sensor. The method includes the following steps: Acquire the first sensing data collected by the first type of sensor; When the first sensing data meets the preset wake-up conditions, the second type of sensor is woken up; Acquire second sensing data collected by the awakened second type of sensor, wherein the second sensing data includes at least image data collected by the camera sensor; Based on the second sensing data, the wearing status parameters of the smart glasses are determined, wherein the wearing status parameters include wearing posture parameters and frame deformation parameters. The wearing posture parameters are determined at least in part based on the position of eye feature points in the image data, and the frame deformation parameters are determined at least in part based on the frame outline shape in the image data.

2. The wearing status detection method as described in claim 1, characterized in that, The first type of sensor includes an ambient light sensor and a capacitance sensor. The first sensing data includes ambient light intensity and capacitance value. After the step of acquiring the first sensing data collected by the first type of sensor, the method further includes: If the ambient light intensity is greater than or equal to a preset light-sensing threshold, and the capacitance value is not within the preset normal capacitance threshold range, then the first sensing data is determined to meet the preset wake-up condition. If the ambient light intensity is less than the preset light-sensing threshold, or the capacitance value falls within the preset normal capacitance threshold range, then it is determined that the first sensing data does not meet the preset wake-up condition.

3. The wearing status detection method as described in claim 2, characterized in that, The second type of sensor also includes motion sensors and time-of-flight sensors. The step of waking up the second type of sensor includes: The motion sensor is activated when the first sensing data meets the preset wake-up condition. The motion sensor acquires motion data, determines the current head posture based on the motion data, and wakes up the camera sensor and the time-of-flight sensor when the deviation of the current head posture from the preset reference head posture is greater than or equal to a preset posture threshold.

4. The wearing status detection method as described in claim 3, characterized in that, The step of determining the wearing status parameters of the smart glasses based on the second sensing data includes: When the wearing status parameters include wearing posture parameters, image data collected by the camera sensor, motion data collected by the motion sensor, and distance data collected by the time-of-flight sensor are acquired. The distance data includes front temple distance data and side temple distance data. The front temple distance data is collected by a time-of-flight sensor located at the frame beam, and the side temple distance data is collected by a time-of-flight sensor located at the temple. Eye feature data is extracted based on the image data, and head posture data is extracted based on the motion data; The wearing posture parameters are generated based on the eye feature data, the head posture data, and the distance data.

5. The wearing status detection method as described in claim 4, characterized in that, The wearing posture parameters include the wearing posture standard. The step of generating the wearing posture parameters based on the eye feature data, the head posture data, and the distance data includes: Calculate the first deviation between the eye feature data and the preset benchmark eye feature, the second deviation between the head posture data and the preset benchmark head posture, and the third deviation between the distance data and the preset benchmark distance; By combining the first deviation, the second deviation, and the third deviation, a comprehensive posture deviation is obtained. The wearing posture standard is obtained based on the posture comprehensive deviation, wherein the wearing posture standard is negatively correlated with the posture comprehensive deviation.

6. The wearing status detection method as described in claim 3, characterized in that, The step of determining the wearing status parameters of the smart glasses based on the second sensing data includes: When the wearing status parameters include the frame deformation parameters, the image data collected by the camera sensor, the distance data collected by the time-of-flight sensor, and the capacitance distribution data collected by the capacitance sensor are acquired. The capacitance distribution data is a dataset formed by arranging the capacitance values ​​output by multiple capacitance sensors deployed at different locations in a preset spatial order. The frame outline data is extracted from the image data, and the temple lateral distance data is extracted from the distance data, wherein the temple lateral distance data is collected by a time-of-flight sensor deployed at the temple position; The frame deformation parameters are generated based on the frame contour data, the temple lateral distance data, and the capacitance distribution data.

7. The wearing status detection method as described in claim 6, characterized in that, The frame deformation parameters include the degree of deformation. The step of generating the frame deformation parameters based on the frame contour data, the temple lateral distance data, and the capacitance distribution data includes: Calculate the fourth deviation between the frame contour data and the preset reference frame contour, the fifth deviation between the temple lateral distance data and the preset reference temple lateral distance, and the sixth deviation between the capacitance distribution data and the preset reference capacitance distribution data; By combining the fourth deviation, the fifth deviation, and the sixth deviation, a comprehensive shape deviation is obtained. The degree of deformation is determined based on the overall shape deviation, wherein the degree of deformation is positively correlated with the overall shape deviation.

8. The wearing status detection method according to any one of claims 3 to 7, characterized in that, Before the step of acquiring the first sensing data collected by the first type of sensor, the method further includes Collect baseline data under normal wearing conditions; Based on the reference data, the light threshold corresponding to the ambient light sensor, the capacitance threshold range and reference capacitance distribution data corresponding to the capacitance sensor, the reference head posture corresponding to the motion sensor, the reference eye features and reference frame outline corresponding to the camera sensor, and the reference distance corresponding to the time-of-flight sensor are respectively set, wherein the reference distance includes the reference frame front distance and the reference temple side distance; After calibration, the motion sensor, the camera sensor, and the time-of-flight sensor are put into a sleep state.

9. A type of smart glasses, characterized in that, The smart glasses include a first type of sensor, a second type of sensor, and a processor, wherein the power consumption of the first type of sensor is lower than that of the second type of sensor; The first type of sensor and the second type of sensor are used to collect sensing data; The processor is configured to perform the steps of the wearing status detection method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a device control program, which, when executed by a processor, implements the steps of the wearing status detection method as described in any one of claims 1 to 8.

Images