A method and system for detecting a wearing state of a smart device, a smart device, a storage medium and a program product

Abstract

The present specification provides a smart device wearing state detection method and system, a smart device, a storage medium and a program product, comprising: collecting first sensing data reflecting a wearing state by a first sensing module embedded in the smart device at a first power consumption level; if the dynamic change amount of the first sensing data within a preset observation time window is within a preset fuzzy interval, it is determined that the current state is a confidence fuzzy state, and the second sensing module embedded in the smart device is activated to collect the pose sequence of the smart device in space at a second power consumption level, the second power consumption level being higher than the first power consumption level; determining the spatial motion trajectory of the smart device based on the pose sequence, and extracting the current geometric feature in the spatial motion trajectory, the current geometric feature being used to reflect the target joint physiological constraint of the wearer of the smart device; if the current geometric feature matches the preset physiological geometric feature, it is determined that the smart device is in a wearing state.

Description

Technical Field

[0001] This specification relates to the field of smart device technology, specifically to a method, system, smart device, storage medium, and program product for detecting the wearing status of a smart device. Background Technology

[0002] Smart wearable devices, such as smart glasses, smartwatches, and smart headsets, typically need to automatically switch their operating modes based on whether the user is wearing the device—performing preset functions when worn and entering a low-power standby state when removed. Accurate and rapid detection of the device's wearing status is a prerequisite for achieving this automatic switching function.

[0003] Currently, the industry standard solution involves incorporating low-power sensing modules into the device and continuously monitoring the signal output by these modules to determine the wearing status. Taking capacitive sensing modules as an example, the basic principle is as follows: capacitive sensing electrodes are arranged on the inner surface of the device that contacts the skin. When the device is worn, a stable coupling capacitance is formed between the electrodes and the skin, maintaining the signal value near a relatively constant baseline. When the device is removed, this coupling capacitance disappears or decreases significantly, causing a signal value transition. By monitoring the signal amplitude or rate of change, it can be determined whether the device is being worn. Besides capacitive sensing modules, existing technologies also employ other low-power sensing modules such as infrared proximity sensors and piezoelectric sensors. Their common characteristic is low power consumption (typically at the microwatt level), making them suitable for long-term operation as normally open detection channels.

[0004] However, these low-power sensing modules share a common technical limitation: the physical quantities they sense are indirect representations of the wearing state, rather than the wearing state itself. Taking capacitive sensing modules as an example, the measurements from the capacitive sensing electrodes are affected not only by the proximity or distance of the human body but also by significant interference from environmental factors such as temperature, humidity, and surface contaminants. When the signal drift caused by environmental factors is similar in amplitude and rate to the actual change in wearing state, the signal output by the sensing module will fall into a fuzzy range—within this range, the signal alone cannot reliably distinguish between "slow drift caused by environmental interference" and "a change in wearing state." Infrared proximity sensing modules face similar problems; changes in ambient light and differences in reflectivity of obstructions can all lead to misjudgments. Therefore, relying solely on these low-power sensing modules for wearing state detection often fails to guarantee accuracy in complex environments, resulting in a significantly higher misjudgment rate.

[0005] Therefore, while existing low-power sensing modules consume little power, they are susceptible to environmental interference because the physical quantities they sense are only indirect representations of the wearing status. This makes it difficult to independently achieve high-accuracy wearing status detection. How to effectively improve the accuracy of wearing status detection while maintaining low power consumption is a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0006] This specification provides a method, system, smart device, storage medium, and program product for detecting the wearing status of a smart device, aiming to solve the technical problems of insufficient detection accuracy of low-power sensing modules in the prior art and the difficulty in achieving high accuracy in wearing status detection while maintaining low power consumption.

[0007] According to a first aspect of the embodiments of this specification, a method for detecting the wearing status of a smart device is provided, comprising: First sensing data reflecting the wearing status is collected at a first power consumption level through a first sensing module embedded in the smart device. If the dynamic change of the first sensing data within a preset observation time window is within a preset fuzzy range, it is determined that the current state is in a confidence fuzzy state, and the second sensing module embedded in the smart device is activated. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level. The spatial motion trajectory of the smart device is determined based on the pose sequence, and the current geometric features in the spatial motion trajectory are extracted. These current geometric features are used to reflect the target joint physiological constraints of the wearer of the smart device. If the current geometric features match the preset physiological geometric features, it is determined that the smart device is in a wearing state.

[0008] According to a second aspect of the embodiments of this specification, a wearing status detection system for a smart device is provided, comprising: The dynamic monitoring unit is configured to collect first sensing data reflecting the wearing status at a first power consumption level through a first sensing module embedded in the smart device. The confidence triggering unit is configured to determine that the current state is confidence ambiguity when the dynamic change of the first sensing data within a preset observation time window is within a preset ambiguity range, and to control the activation of the second sensing module embedded in the smart device. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level. The feature extraction unit is configured to determine the spatial motion trajectory of the smart device based on the pose sequence, and extract the current geometric features in the spatial motion trajectory, which are used to reflect the target joint physiological constraints of the wearer of the smart device. The status confirmation unit is configured to determine that the smart device is in a wearing state when the current geometric features match the preset physiological geometric features.

[0009] According to a third aspect of the embodiments of this specification, a smart device is provided, including a first sensing module, a second sensing module, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the first sensing module acquires first sensing data reflecting a wearing state at a first power consumption level, the second sensing module acquires a pose sequence of the smart device in space at a second power consumption level, the second power consumption level being higher than the first power consumption level, and the processor implements the method described in the first aspect when executing the computer program.

[0010] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the method described in the first aspect above.

[0011] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, including computer program instructions that, when executed by a processor, implement the method described in the first aspect above.

[0012] The technical solution provided in the embodiments of this specification allows the first sensing module to operate continuously with low power consumption, undertaking the task of routine wear status monitoring. When the change in the first sensing data falls within a preset fuzzy range, it indicates that the first sensing module cannot reliably distinguish between environmental interference and actual wear status changes. Only then is the second sensing module, which has higher power consumption, activated for secondary verification. Since the fuzzy range occurs far less frequently than in normal use, the cumulative operating time of the second sensing module is extremely low, enabling the device to maintain an overall average power consumption level close to that of the first sensing module operating alone.

[0013] After the second sensing module is activated, it reconstructs the device's motion trajectory in three-dimensional space by collecting the device's pose sequence in space, and extracts geometric features reflecting the physiological constraints of the wearer's target joints. Since the physiological constraints of human joints are inherent physical properties at the anatomical level and are not affected by factors such as environmental temperature and humidity, the matching judgment based on these geometric features has a natural anti-interference capability, which can effectively solve the detection reliability problem of the first sensing module in the fuzzy range, thereby achieving high accuracy in wearing status detection while maintaining low power consumption. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the structure and module deployment of a smart device provided in the embodiments of this specification.

[0015] Figure 2 This is a flowchart of a wearing status detection method provided in the embodiments of this specification.

[0016] Figure 3 This is a schematic diagram of a preset fuzzy interval and state value boundary provided in the embodiments of this specification.

[0017] Figure 4 This is a schematic diagram of the geometric constraints of head rotation in a wearing state provided in the embodiments of this specification.

[0018] Figure 5 This is a timing diagram of sensor activation and sleep states provided in the embodiments of this specification.

[0019] Figure 6 This is a flowchart of pose sequence acquisition data processing provided in the embodiments of this specification.

[0020] Figure 7 This is a schematic diagram of a wearing status detection system provided in the embodiments of this specification.

[0021] Figure 8 This is a schematic diagram of the hardware structure of a smart device provided in the embodiments of this specification. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0023] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this manual are all information and data authorized by the user or fully authorized by all parties. The collection, use and processing of related data shall comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals shall be provided for users to choose to authorize or refuse.

[0024] The wearing status detection method provided in the embodiments of this specification can be applied to various smart devices. In practical applications, the smart device is usually an electronic device integrated into a wearable carrier. Depending on the wearing position, the wearable carrier may include: upper limb devices such as bracelets, watches, gloves, and armbands; lower limb devices such as pants, shoes, insoles, and leg bands; torso devices such as clothing, necklaces, brooches, and exoskeletons; and head-mounted devices such as glasses, helmets, headphones, headbands, and hats. This specification does not limit this. These smart devices typically rely on accurate wearing status detection to trigger corresponding device functions, such as deciding whether to activate the display module, enable voice interaction, or perform authentication operations.

[0025] Existing smart device wear detection solutions typically require a balance between standby power consumption and detection accuracy. Taking AR (Augmented Reality) glasses as an example, a common approach is to use capacitive sensors embedded in the temples for detection. This solution has low power consumption but is easily affected by changes in ambient temperature and humidity. When a user moves from indoors to outdoors, or when sweating increases skin moisture, the capacitance value drifts, easily causing false triggers or missed detections. On the other hand, some devices incorporate cameras and IMUs (Inertial Measurement Units) for pose estimation, but this is mainly used for AR positioning or eye tracking and is not effectively used for wear status detection. Furthermore, pose estimation has high power consumption and latency; if it is constantly running, it will significantly shorten the device's battery life.

[0026] To balance low power consumption and detection accuracy, this specification provides a graded wake-up detection method. This method primarily relies on low-power sensors for monitoring under normal conditions, only activating higher-power sensors for auxiliary judgment when signal uncertainty arises, thus rationally allocating system power consumption. The system structure provided in this specification's embodiments will now be described with reference to the accompanying drawings. For ease of understanding, subsequent embodiments will use smart glasses as an example smart device. Figure 1This diagram illustrates the structure and module deployment of a smart device 100 provided in an embodiment of this specification. In one embodiment, the smart device 100 takes the form of smart glasses, and its structure includes a frame, temples, and nose pads. The smart device 100 integrates sensing components with different power consumption levels, specifically including a first sensing module 101, a second sensing module 102, a processor 103, and a memory 104. The first sensing module 101 is typically deployed on the inner side of the temple in contact with the skin, used to continuously sense changes in physical quantities caused by the approach or contact of a person with low power consumption. The second sensing module 102 is deployed inside the temple or frame, used to collect spatial pose data under specific conditions. The processor 103 is connected to the first sensing module 101, the second sensing module 102, and the memory 104, respectively, and is used to execute computer program instructions stored in the memory 104. Figure 1 The hardware structure shown is as follows: Figure 2 The flowchart of the wearing status detection method provided in the embodiments of this specification is shown, which specifically includes the following steps.

[0027] Step 201: First sensing data reflecting the wearing status is collected at a first power consumption level by a first sensing module embedded in the smart device.

[0028] In this step, the first sensing module 101 is in a continuously operating state, and its main design goal is to maintain low standby power consumption. The first power consumption level typically refers to the average power consumption of the sensor in continuous sampling or low-frequency polling mode, to ensure that the smart device 100 does not consume too much power during a long standby period. The first sensing data is essentially the electrical signal or digital quantized value generated when the first sensing module 101 physically couples with the surrounding environment or human tissue.

[0029] In one specific implementation, the first sensing module 101 can employ a capacitive sensing electrode array. When a user picks up the smart glasses and brings them close to their head, the human skin, acting as a conductive medium, enters the edge electric field range of the capacitive electrodes, causing a change in the equivalent parasitic capacitance between the electrodes and ground. The processor 103 reads this capacitance value through an internal capacitance-to-digital converter, forming a first sensing data sequence. This detection method based on electric field coupling does not require active emission of a light source or radio frequency signal, offering the advantage of low power consumption. In another different embodiment, the first sensing module 101 can also be replaced with an infrared proximity sensor or a micro-mechanical switch. If an infrared proximity sensor is used, it generates distance data by emitting low-duty-cycle infrared pulses and receiving reflected light; if a micro-mechanical switch is used, a closing signal is triggered by physical deformation at the temple hinge, providing a mechanical wearing indication. Regardless of the physical medium used, the first sensing module 101 is used to maintain environmental condition monitoring with low power consumption.

[0030] Step 202: If the dynamic change of the first sensing data within the preset observation time window is within the preset fuzzy range, it is determined that the current state is in a confidence fuzzy state, and the second sensing module embedded in the smart device is activated. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level.

[0031] This step is primarily used to determine whether the second sensing module 102 needs to be activated. The dynamic change here refers to a quantified parameter reflecting the degree of fluctuation of the sensing signal within a preset time window. Its calculation methods can include the rate of change, variance, or frequency domain energy distribution. The preset observation time window is a sliding or fixed time interval used to smooth out instantaneous noise. When the dynamic change is low, it indicates that the signal is in a stable state; when the dynamic change is high, it indicates that significant physical contact or separation has occurred; and when the dynamic change is in the middle, it means that the signal characteristics fall into the region where environmental interference and actual action overlap. The confidence ambiguity state here refers to the intermediate state where the system determines that the current signal has uncertainty (i.e., an "uncertain" state) and requires secondary verification.

[0032] In a simple implementation, the system can set a single confidence threshold. However, this approach has limitations: when the smart device is in a stable wearing state, because the human body and the device are relatively stationary, the dynamic change in the capacitance signal will approach zero, which may cause the system to misjudge "stable wearing" as "confidence ambiguity." To avoid this misjudgment, this embodiment refines the preset ambiguity interval. Specifically, the preset ambiguity interval is the numerical range between a preset first confidence threshold and a preset second confidence threshold, where the first confidence threshold is lower than the second confidence threshold. When the dynamic change is lower than the first confidence threshold, the first sensor data indicates that the smart device is in a stable wearing state; when the dynamic change is higher than the second confidence threshold, the first sensor data indicates that the smart device is in a transitional wearing / removing state.

[0033] By introducing a first confidence threshold and a second confidence threshold, the system divides the numerical range into three state intervals. Figure 3 The preset ambiguity interval and the numerical boundaries of the three states are illustrated using a number line, clearly marking the first confidence threshold 301, the second confidence threshold 302, and the preset ambiguity interval 303 between them. When the dynamic change is below the first confidence threshold 301, the system considers the signal to be in a stable period; when the dynamic change is above the second confidence threshold 302, the signal exhibits obvious jump characteristics; only when the dynamic change is between the two does the signal exhibit slow change characteristics, which is usually caused by baseline drift due to fluctuations in ambient temperature and humidity or slight friction. At this time, the system determines that it is currently in a confidence ambiguity state and sends an activation command to the second sensing module 102.

[0034] In the above embodiments, the first confidence threshold and the second confidence threshold can be preset fixed values. However, as the device is used for longer periods, the device characteristics may drift. To adapt the ambiguity interval to the current usage environment, this embodiment provides an adaptive calibration method. Specifically, the preset ambiguity interval can be defined as follows: collecting the statistical distribution of the first sensor data within a preset historical time period; determining the center value of the preset ambiguity interval based on the central trend of the statistical distribution; and determining the half-width of the preset ambiguity interval based on the dispersion of the statistical distribution, thereby obtaining the upper and lower bounds of the preset ambiguity interval. Under this adaptive calibration mechanism, the processor 103 records the first sensor data of the device in a stable state in the background and constructs its probability density distribution model. The central trend of the statistical distribution (such as the mean) reflects the baseline level in the current environment, while the dispersion of the statistical distribution (such as the standard deviation) quantifies the fluctuation amplitude of environmental noise. For example, in an outdoor environment with large temperature and humidity changes, the dispersion increases, and the system automatically widens the upper and lower bounds of the ambiguity interval. This adaptive threshold algorithm dynamically adjusts the capacitance determination boundary based on historical wearing data and supports personalized calibration when the user wears the device for the first time.

[0035] Based on the above embodiments, the first sensing module and the second sensing module can employ various types of sensors. Specifically, the first sensing module can be a capacitive sensing module, and / or the second sensing module can be an image acquisition unit and / or an inertial measurement unit. Figure 6 The data processing flow for pose sequence acquisition by the second sensing module is illustrated, specifically the implementation flow of the VINS (Visual-Inertial Navigation System) algorithm. This includes an image acquisition unit 601, an inertial acquisition unit 602, and a fusion calculation module 603. When the second sensing module simultaneously includes an image acquisition unit and an inertial measurement unit, the system can acquire accelerometer and gyroscope data through the inertial acquisition unit 602 and environmental image data through the image acquisition unit 601. Subsequently, the fusion calculation module 603 applies the VINS algorithm based on these two types of data to estimate the three-dimensional attitude angles and position sequence of the intelligent device in space. This multimodal fusion method can utilize visual data to correct the accumulated errors of inertial data in scenes with rich environmental textures, thereby improving the accuracy of pose estimation.

[0036] Step 203: Determine the spatial motion trajectory of the smart device based on the pose sequence, and extract the current geometric features in the spatial motion trajectory. These current geometric features are used to reflect the target joint physiological constraints of the wearer of the smart device.

[0037] After acquiring the pose sequence, the system needs to extract features that reflect the wearing status. When a smart device is worn correctly, its movement is usually physically restricted by specific joints of the human body (such as the cervical spine or wrist joint). This physiological constraint of the target joint causes the spatial movement trajectory of the device to exhibit specific geometric patterns.

[0038] In practical use, if the original pose sequence is used directly, the overall displacement of the torso will be superimposed on the head movement when the wearer is walking or running, causing translational components unrelated to joint rotation to be mixed into the extracted geometric features. To eliminate this interference, this embodiment performs trajectory preprocessing on the pose sequence before extracting geometric features to remove displacement components unrelated to the wearer's joint movements, resulting in a purified pose sequence. Accordingly, the spatial motion trajectory of the smart device is determined based on the pose sequence, and the current geometric features in the spatial motion trajectory are extracted. Specifically, this includes: determining the purified spatial motion trajectory based on the purified pose sequence and extracting the current geometric features in the purified spatial motion trajectory.

[0039] Furthermore, considering that the displacement components unrelated to joint movement in the pose sequence have different characteristics under different exercise intensities, the system generates the current exercise intensity level based on the motion state quantification characteristics of the pose sequence. In a specific embodiment, when the average linear velocity is less than 0.8 m / s and the standard deviation of acceleration is less than 1.5 m / s², it is determined to be low-speed exercise (such as sitting, standing, or walking slowly); when the average linear velocity is greater than or equal to 0.8 m / s, or the standard deviation of acceleration is greater than or equal to 1.5 m / s², it is determined to be high-speed exercise (such as walking briskly, running, or jumping).

[0040] For different motion intensity levels, this embodiment provides two specific trajectory preprocessing methods within a preset time window of 1 to 2 seconds. Specifically, when the current motion intensity level is high-speed, a sliding window de-trending method is used to remove low-frequency trend terms from the pose sequence within the preset time window; and / or, when the current motion intensity level is low-speed, local trajectory fitting is performed on the pose sequence within the preset time window. During high-speed motion, the low-frequency drift introduced by torso displacement is significant, and the sliding window de-trending method can effectively remove these global translation baseline components. However, during low-speed motion, torso displacement is minimal; if the de-trending method is still used, it may over-clean the data, erasing minute real rotation signals. Therefore, local trajectory fitting is used to retain more high-frequency details.

[0041] After the above preprocessing, the system can extract specific geometric features from the trajectory of motion in the clean space, such as calculating the instantaneous radius of curvature. . Figure 4 A schematic diagram illustrating the geometric constraints of head rotation when the device is worn is shown. Figure 4 As shown, when the smart device 403 is worn on the head, its movement is constrained by the cervical joints, and its center of rotation in space is relatively fixed (i.e., near the cervical pivot 401). Therefore, the spatial motion trajectory 402 of the smart device 403 as the head rotates exhibits a specific arc characteristic, and the instantaneous radius of curvature R between the smart device 403 and the cervical pivot 401 (equivalent to the aforementioned...) The movement of the head is strictly limited by the physiological structure of the human body. This physiological constraint is reflected in various head movements: for example, nodding (pitch) involves rotation around the horizontal axis, shaking the head (yaw) involves rotation around the vertical axis, and tilting (roll) involves rotation around the front-back axis. These movements are all approximately equivalent to the rotation of a rigid body around the cervical vertebral pivot 401 at a fixed point. In contrast, when a smart device is not worn (e.g., placed on a table, shaken freely in the hand, or dropped), its movement often exhibits irregular random motion, mainly translational, with an unstable center of rotation, and may even show high-frequency shaking or free-fall characteristics. Therefore, using cervical rotation constraint as a criterion for wearing the device is highly physically feasible.

[0042] Step 204: If the current geometric features match the preset physiological geometric features, determine that the smart device is in a wearing state.

[0043] This step primarily outputs the final wearing determination result based on the extracted geometric features. Corresponding to the different geometric feature types mentioned above, this embodiment provides specific determination methods for matching preset physiological geometric features. Specifically, the preset physiological geometric features are geometric feature intervals defined by the anatomical constraints of the target joint. If the current geometric feature includes the instantaneous radius of curvature... If the current geometric feature matches the preset physiological geometric feature, including the instantaneous radius of curvature, then the current geometric feature matches the preset physiological geometric feature. The current geometric feature remains within a preset radius of curvature (i.e., within the neck rotation radius range). If the current geometric feature includes a relative rotation center component, then the current geometric feature matches a preset physiological geometric feature, including the relative rotation center component remaining within the preset rotation center range. Through this white-box determination method based on physical constraints, the system can effectively distinguish whether the device is worn on the human body or is randomly placed or shaken by hand, thereby reducing the false positive rate.

[0044] To further suppress the jitter in the instantaneous judgment result, this embodiment will use the above-mentioned instantaneous radius of curvature as a basis. The matching results are input to a first-order hysteresis filter for smoothing, and the final wearing status determination result is output. The first-order hysteresis filter can effectively filter out glitches caused by brief atypical head movements or sensor noise, ensuring the stability of the status output.

[0045] After introducing filtering and continuity verification, the reliability of the judgment within a single activation cycle is improved. However, when the dynamic change of the first sensing data fluctuates repeatedly near the boundary of the preset ambiguity interval, the second sensing module may be frequently activated, leading to increased system power consumption. To suppress this high-frequency wake-up phenomenon, this embodiment further introduces a sleep cycle mechanism. Specifically, after completing one auxiliary judgment, the second sensing module (i.e., the vision-inertial subsystem) forcibly executes a sleep cycle; during this sleep cycle, the activation state of the second sensing module is independent of whether the dynamic change of the first sensing data is within the preset ambiguity interval. Figure 5 The timing relationship between the activation and sleep states of the second sensing module is illustrated in the form of a timing waveform diagram. As shown in the figure, when the dynamic change of the first sensing data crosses the boundary of the fuzzy interval for the first time, it precisely corresponds to the start of the activation period 501 when the state of the second sensing module jumps to a high level. After completing one acquisition, the module immediately enters the low-level sleep period 502. During the sleep period 502, even if the upper signal exhibits high-frequency, dense, and repetitive fluctuations near the boundary, the lower state is still forced to maintain a flat low level, effectively shielding the frequent start-stop caused by boundary jitter and further optimizing the overall power consumption of the system.

[0046] The wearing status detection method provided in the embodiments of this specification can be applied to various practical scenarios. Taking the application of smart glasses in security scenarios such as mobile payment or identity verification as an example, accurate wearing detection is an important prerequisite for preventing others from stealing the device or accidentally triggering the payment process. In these scenarios, this solution can effectively distinguish whether the device is worn on the user's head or held by someone else through graded wake-up and physiological geometric feature verification. For example, when the user takes off the glasses and hands them to the cashier, the capacitance signal changes drastically, and the system determines that it is a transition from wearing to removing the glasses and immediately locks the payment function; if the ambient humidity is high, causing the capacitance to drift into the fuzzy range, the system activates the second sensing module, and by extracting the radius of curvature of the neck, it finds that it does not meet the neck constraints (such as the glasses being placed flat on a table), thus confirming that it is not being worn and keeping the payment locked. At this time, the system can automatically pop up a password input interface on the lens display screen, or call the bound smartphone for secondary fingerprint / facial biometric verification to ensure the security of funds. At the same time, in high-frequency interaction scenarios such as daily voice interaction or information prompts, such as when the user is running outdoors (high-speed exercise), the first sensing module may drift into the fuzzy range due to sweating. After the system activates the second sensing module, it uses a sliding window de-trending method to remove the low-frequency trend of torso translation, extracts the pure head rotation curvature radius, and confirms the wearing status. This allows users to smoothly wake up the smart assistant via voice, alleviating users' battery anxiety and interaction interruption issues.

[0047] Based on the same inventive concept as the method embodiments, this specification also provides a smart device wearing status detection system. Figure 7 A schematic diagram of the system architecture corresponding to this deployment scenario is shown. This system can be deployed in the aforementioned smart device, serving as the execution entity for each method step. Specifically, the system includes: The dynamic monitoring unit 701 is configured to collect first sensing data reflecting the wearing status at a first power consumption level through a first sensing module embedded in the smart device.

[0048] The confidence trigger unit 702 is configured to determine that the current state is confidence ambiguity when the dynamic change of the first sensing data within a preset observation time window is within a preset ambiguity range, and to control the activation of the second sensing module embedded in the smart device. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level.

[0049] The feature extraction unit 703 is configured to determine the spatial motion trajectory of the smart device based on the pose sequence and extract the current geometric features in the spatial motion trajectory. The current geometric features are used to reflect the target joint physiological constraints of the wearer of the smart device.

[0050] The status confirmation unit 704 is configured to determine that the smart device is in a wearing state when the current geometric features match the preset physiological geometric features.

[0051] Optionally, the preset fuzzy range is the numerical range between a preset first confidence threshold and a preset second confidence threshold, where the first confidence threshold is lower than the second confidence threshold. When the dynamic change is lower than the first confidence threshold, the first sensor data indicates that the smart device is in a stable wearing state. When the dynamic change is higher than the second confidence threshold, the first sensor data indicates that the smart device is in a transitional wearing / removing state.

[0052] Optionally, the system further includes: a trajectory preprocessing unit 705, used to perform trajectory preprocessing operations on the pose sequence before extracting the current geometric features in the spatial motion trajectory, to remove displacement components unrelated to the wearer's joint movements, thereby obtaining a clean pose sequence. The feature extraction unit 703 is specifically used to: determine the clean spatial motion trajectory based on the clean pose sequence, and extract the current geometric features in the clean spatial motion trajectory.

[0053] Optionally, the trajectory preprocessing unit 705 is specifically used to: generate the current motion intensity level based on the motion state quantization features of the pose sequence; and perform trajectory preprocessing operations corresponding to the current motion intensity level on the pose sequence according to the current motion intensity level.

[0054] Optionally, the trajectory preprocessing unit 705 is specifically used to: remove low-frequency trend terms from the pose sequence within a preset time window by using a sliding window detrending method when the current motion intensity level is a high-speed motion level; and / or, perform local trajectory fitting on the pose sequence within a preset time window when the current motion intensity level is a low-speed motion level.

[0055] Optionally, the current geometric features extracted by the feature extraction unit 703 include the instantaneous radius of curvature and / or the relative rotation center component of the spatial motion trajectory in the three-dimensional spatial projection.

[0056] Optionally, the state confirmation unit 704 is specifically used for: setting the preset physiological geometric features as a geometric feature range defined by the anatomical constraints of the target joint; if the current geometric feature includes an instantaneous radius of curvature, then the current geometric feature matches the preset physiological geometric features, including maintaining the instantaneous radius of curvature within the preset radius of curvature; if the current geometric feature includes a relative rotation center component, then the current geometric feature matches the preset physiological geometric features, including maintaining the relative rotation center component within the preset rotation center range.

[0057] Optionally, the status confirmation unit 704 is specifically used to: count the number of consecutive effective cycles in which the current geometric feature matches the preset physiological geometric feature; and determine that the smart device is in the wearing state when the number of consecutive effective cycles reaches the preset continuity threshold.

[0058] Optionally, the confidence trigger unit 702 is also used to control the second sensing module to forcibly execute a sleep cycle after completing the acquisition of a pose sequence; during the sleep cycle, the activation state of the second sensing module is independent of whether the dynamic change of the first sensing data is within a preset fuzzy range.

[0059] Optionally, the preset fuzzy interval is defined as follows: the statistical distribution of the first sensor data collected within a preset historical period is obtained; the center value of the preset fuzzy interval is determined based on the central trend of the statistical distribution, and the half-width of the preset fuzzy interval is determined based on the dispersion of the statistical distribution, so as to obtain the upper and lower bounds of the preset fuzzy interval.

[0060] Optionally, the first sensing module is a capacitive sensing module, and / or the second sensing module is an image acquisition unit and / or an inertial measurement unit.

[0061] This specification also provides an example of a smart device. Figure 8 A schematic diagram of the hardware structure of the smart device is shown. The smart device includes a processor 801, a memory 802, a communication interface 803, a bus 804, and a first sensing module 805 (equivalent to...). Figure 1 The first sensing module 101) and the second sensing module 806 (equivalent to Figure 1The second sensing module 802). The first sensing module 805 collects first sensing data reflecting the wearing state at a first power consumption level, and the second sensing module 806 collects the pose sequence of the smart device in space at a second power consumption level, where the second power consumption level is higher than the first power consumption level. When the processor 801 executes the computer program stored in the memory 802, it implements the steps in any of the aforementioned method embodiments. The processor 801, memory 802, communication interface 803, first sensing module 805, and second sensing module 806 all interact via bus 804.

[0062] Based on the same concept as the methods described above, this specification also provides a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0063] Based on the same concept as the methods described above, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0064] What those skilled in the art will understand is: In this specification, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus 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, product, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded.

[0065] In this specification, “a,” “an,” and “the” do not specifically refer to the singular, but may also include the plural.

[0066] In this specification, ordinal numbers such as "first," "second," etc., do not necessarily indicate order; they are often used to distinguish between objects. For example, "first server" and "second server" usually refer to two servers. To differentiate between these two servers, they are described as "first server" and "second server." Of course, sometimes these two servers may be the same server.

[0067] In this specification, unless explicitly stated otherwise, "receiving and sending data" does not necessarily mean direct receiving and sending; it can also mean indirect receiving and sending. For example, A receiving data sent by B can be understood as A directly receiving the data sent by B, or it can be understood as A indirectly receiving the data sent by B through other entities such as C. Similarly, B sending data to A can be understood as B sending the data directly to A, or it can be understood as B indirectly sending the data to A through other entities such as C. Here, C can be one entity, or it can be two or more entities.

[0068] In this specification, unless explicitly stated otherwise, the relationships between structures can be direct or indirect. For example, when describing "A is connected to B," unless it is explicitly stated that A and B are directly connected, it should be understood that A can be directly connected to B or indirectly connected to B. Similarly, when describing "A is on top of B," unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B or indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0069] This specification uses specific terms to describe embodiments thereof. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of those different embodiments or examples, without contradiction.

[0070] Although one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is only one of many possible execution orders and does not represent the only execution order. Therefore, when the claims involve method steps, any changes or adjustments to the order of such steps, or the parallelism between steps, are also within the scope of protection of the claims.

Claims

1. A method for detecting the wearing status of a smart device, comprising: First sensing data reflecting the wearing status is collected at a first power consumption level through a first sensing module embedded in the smart device. If the dynamic change of the first sensing data within a preset observation time window is within a preset fuzzy range, it is determined that the current state is in a confidence fuzzy state, and the second sensing module embedded in the smart device is activated. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level. The spatial motion trajectory of the smart device is determined based on the pose sequence, and the current geometric features in the spatial motion trajectory are extracted. These current geometric features are used to reflect the target joint physiological constraints of the wearer of the smart device. If the current geometric features match the preset physiological geometric features, it is determined that the smart device is in a wearing state.

2. The method according to claim 1, wherein, The preset fuzzy interval is the numerical range between a preset first confidence threshold and a preset second confidence threshold, wherein the first confidence threshold is lower than the second confidence threshold; When the amount of dynamic change is lower than the first confidence threshold, the first sensing data indicates that the smart device is in a stable wearing state; If the amount of dynamic change is higher than the second confidence threshold, the first sensing data indicates that the smart device is in a transitional state between wearing and removing the device.

3. The method according to claim 1, wherein, Before extracting the current geometric features in the spatial motion trajectory, the method further includes: A trajectory preprocessing operation is performed on the pose sequence to remove displacement components that are unrelated to the wearer's joint movements, thereby obtaining a purified pose sequence. The step of determining the spatial motion trajectory of the smart device based on the pose sequence and extracting the current geometric features in the spatial motion trajectory includes: determining the purification spatial motion trajectory based on the purification pose sequence and extracting the current geometric features in the purification spatial motion trajectory.

4. The method according to claim 3, wherein, The trajectory preprocessing operation performed on the pose sequence includes: Based on the motion state quantification features of the pose sequence, the current motion intensity level is generated; Based on the current motion intensity level, perform trajectory preprocessing operations on the pose sequence corresponding to the current motion intensity level.

5. The method according to claim 4, wherein, The step of performing trajectory preprocessing operations on the pose sequence corresponding to the current motion intensity level, based on the current motion intensity level, includes: When the current motion intensity level belongs to the high-speed motion level, a sliding window detrending method is used to remove low-frequency trend terms from the pose sequence within a preset time window; and / or When the current motion intensity level is low-speed motion level, local trajectory fitting is performed on the pose sequence within the preset time window.

6. The method according to claim 1, wherein, The current geometric features include the instantaneous radius of curvature and / or the relative rotation center component of the spatial motion trajectory in the three-dimensional spatial projection.

7. The method according to claim 6, wherein, The preset physiological geometric features are geometric feature ranges defined by the anatomical constraints of the target joint; If the current geometric feature includes the instantaneous radius of curvature, then the current geometric feature matches the preset physiological geometric feature, including the instantaneous radius of curvature being maintained within the preset radius of curvature. If the current geometric feature includes the relative rotation center component, then the current geometric feature matches the preset physiological geometric feature, including the relative rotation center component being maintained within the preset rotation center range.

8. The method according to claim 1, wherein, The step of determining that the smart device is in a wearing state when the current geometric features match preset physiological geometric features includes: The number of consecutive effective cycles in which the current geometric feature matches the preset physiological geometric feature is counted; When the number of consecutive effective cycles reaches a preset continuity threshold, it is determined that the smart device is in a wearing state.

9. The method according to claim 1, wherein, After completing the acquisition of one pose sequence, the second sensing module forcibly executes a sleep cycle; During the sleep cycle, the activation state of the second sensing module is independent of whether the dynamic change of the first sensing data is within the preset fuzzy range.

10. The method according to claim 1, wherein, The preset fuzzy interval is defined in the following manner: Collect the statistical distribution of the first sensor data within a preset historical time period; The center value of the preset fuzzy interval is determined based on the central tendency of the statistical distribution, and the half-width of the preset fuzzy interval is determined based on the dispersion of the statistical distribution, so as to obtain the upper and lower bounds of the preset fuzzy interval.

11. The method according to claim 1, wherein, The first sensing module is a capacitive sensing module, and / or the second sensing module is an image acquisition unit and / or an inertial measurement unit.

12. A wearing status detection system for a smart device, comprising: The dynamic monitoring unit is configured to collect first sensing data reflecting the wearing status at a first power consumption level through a first sensing module embedded in the smart device. The confidence triggering unit is configured to determine that the current state is confidence ambiguity when the dynamic change of the first sensing data within a preset observation time window is within a preset ambiguity range, and to control the activation of the second sensing module embedded in the smart device. The second sensing module is used to collect the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level. The feature extraction unit is configured to determine the spatial motion trajectory of the smart device based on the pose sequence, and extract the current geometric features in the spatial motion trajectory, which are used to reflect the target joint physiological constraints of the wearer of the smart device. The status confirmation unit is configured to determine that the smart device is in a wearing state when the current geometric features match the preset physiological geometric features.

13. An intelligent device, comprising a first sensing module, a second sensing module, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein, The first sensing module collects first sensing data reflecting the wearing state at a first power consumption level, and the second sensing module collects the pose sequence of the smart device in space at a second power consumption level, wherein the second power consumption level is higher than the first power consumption level, and the processor implements the method as described in any one of claims 1 to 11 when executing the computer program.

14. A computer-readable storage medium having stored thereon computer program instructions, wherein, When the computer program instructions are executed by the processor, they implement the method as described in any one of claims 1 to 11.

15. A computer program product comprising computer program instructions, wherein, When the computer program instructions are executed by the processor, they implement the method as described in any one of claims 1 to 11.

Images