Smart wearable device wearing detection method, control device and smart wearable device

Abstract

The application discloses a kind of smart wearable device wearing detection method, control device and smart wearable device, it is related to smart wearable device technical field, wherein, smart wearable device wearing detection method is verified by successively executing wearing posture, temperature verification, human bioelectricity verification and photoelectric signal verification, and four levels are formed in series verification link, only four verification passes through all to determine normal wearing state, can effectively exclude the situation of single sensor misjudgment, improve the accuracy of wearing detection;Meanwhile, by setting up posture sensor, temperature sensor, electrical sensor and optical sensor four different types of sensors, detect from equipment posture, body temperature, human bioelectricity and blood flow signal four different physical dimensions respectively;The technical scheme provided by the application can solve the problem that the existing smart wearable device is caused by relying on single PPG sensor and the wearing detection misjudgment rate is high.

Description

Technical Field

[0001] This invention relates to the field of smart wearable device technology, and in particular to a smart wearable device wearing detection method, control device, and smart wearable device. Background Technology

[0002] Currently, wear detection for smart wearable devices mainly relies on photoplethysmography (PPG), which involves emitting light signals and receiving reflected light from the skin, then judging the wearing status based on changes in blood flow. However, this method has many problems in practical use: reflected light signals are easily misinterpreted as being worn when the device is placed on a non-human surface such as a table; spurious signals caused by clothing can interfere with the detection results; vasoconstriction in low temperatures weakens blood flow signals, affecting the accuracy of the judgment; artifacts caused by muscle vibration during exercise can also easily lead to misjudgments; and it is difficult to distinguish between different scenarios such as close to the skin, close to a warm non-living surface, and static placement. These factors result in a high misjudgment rate for wear detection, easily distorted health monitoring data, and potential accidental triggering of device functions. Therefore, improving the accuracy of wear detection for smart wearable devices has become an urgent problem to be solved in this field. Summary of the Invention

[0003] The main objective of this invention is to propose a method, control device, and smart wearable device for detecting wearability, aiming to solve the problem of high false detection rate in existing smart wearable devices due to their reliance on a single PPG sensor.

[0004] To achieve the above objectives, the present invention proposes a smart wearable device wearing detection method, which is applied to a smart wearable device equipped with an attitude sensor, an optical sensor, a temperature sensor, and an electrical sensor. The method includes: The posture sensor collects posture data of the smart wearable device and verifies the wearing posture based on the device posture data. If the wearing posture verification is passed, the temperature sensor collects the body surface temperature signal and performs temperature verification based on the body surface temperature signal. If the temperature verification is passed, the electrical sensor collects human skin current signals and performs human bioelectric verification based on the skin current signals. If the bioelectric verification is successful, the optical sensor acquires blood flow signals and performs photoelectric signal verification based on the blood flow signals. If the wearing posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification all pass, it is determined to be a normal wearing state.

[0005] In one embodiment, the posture sensor collects posture data of the smart wearable device, and the step of verifying the wearing posture based on the device posture data includes: The attitude sensor collects acceleration and angular velocity data; The acceleration data and the angular velocity data are filtered. Based on the filtered acceleration data and angular velocity data, the mean acceleration, variance acceleration, rate of acceleration crossing zero, root mean square angular velocity, pitch angle and roll angle of the smart wearable device are extracted as attitude features. The average acceleration is between 0.8g and 1.2g, the pitch angle and roll angle are both within ±15°, and the root mean square angular velocity is less than 10°. o If the wearing posture verification is successful, the result is determined to be / s.

[0006] In one embodiment, the temperature sensor acquires a body surface temperature signal, and the step of verifying the temperature based on the body surface temperature signal includes: The temperature sensor collects body surface temperature data; The body surface temperature data is processed by a moving average filter, and the temperature value is output. If the temperature value is within the range of 32℃ to 37.5℃, the temperature verification is deemed successful.

[0007] In one embodiment, the step of using the electrical sensor to collect human skin current signals and performing human bioelectric verification based on the skin current signals includes: The electrical sensor collects skin current signals and electrocardiogram signals; Calculate the signal-to-noise ratio of the skin current signal and the electrocardiogram signal; If the skin conductance value is between 10 μS and 100 μS and the signal-to-noise ratio of the electrocardiogram signal is greater than 15 dB, the bioelectric verification is deemed to be successful.

[0008] In one embodiment, the optical sensor acquires blood flow signals, and the step of performing photoelectric signal verification based on the blood flow signals includes: The optical sensor collects blood flow signals; Extract the pulse wave peak value and signal-to-noise ratio of the blood flow signal; If the peak value of the pulse wave is greater than 60% of the first preset reference value and the signal-to-noise ratio is greater than 12dB, the photoelectric signal verification is deemed successful.

[0009] In one embodiment, the smart wearable device wearing detection method further includes a re-detection step; If any one of the following verifications fails: wearing posture verification, temperature verification, bioelectric verification, and photoelectric signal verification, the failed verification dimension and the corresponding sensor data shall be recorded. Initiate a delayed re-examination mechanism, re-execute the verification of failed dimensions after a 2-second interval, and adopt a dynamic threshold adjustment strategy during the re-examination process to adjust the current verification threshold to a threshold that is suitable for the current detection scenario; If the threshold is adjusted and the verification is successful, proceed with the subsequent verification steps. If the test fails three times, the device is deemed to be not worn or worn in an invalid state.

[0010] In one embodiment, the dynamic threshold adjustment strategy includes: In motion scenarios, the angular velocity threshold of the attitude sensor is adjusted from 10° / s to 15° / s; In an environment where the ambient temperature is below 10°C, the lower limit of the temperature threshold for temperature verification is adjusted from 32°C to 30°C, and the lower limit of the skin conductance threshold for bioelectric verification is adjusted from 10μS to 8μS. In environments where the ambient temperature is above 35°C, the upper limit of the temperature threshold for temperature verification is adjusted from 37.5°C to 38°C.

[0011] The present invention also proposes a control device, the control device comprising: a memory, a processor, and a control program for detecting the wearing of a smart wearable device stored in the memory and executable on the processor, the control program for detecting the wearing of a smart wearable device being configured to implement the smart wearable device wearing detection method as described in any one of the above.

[0012] The present invention also proposes a smart wearable device, which includes an attitude sensor, an optical sensor, a temperature sensor, an electrical sensor, and a control device as described above.

[0013] In one embodiment, the attitude sensor is disposed inside the smart wearable device, and the attitude sensor is used to collect acceleration data and angular velocity data; The optical sensor is disposed on the surface of the smart wearable device that contacts the skin. The optical sensor is used to collect light signals reflected from the skin and output blood flow signals. The electrical sensors are symmetrically arranged on both sides of the contact surface of the smart wearable device, and the electrical sensors are used to collect the current signal of human skin. The temperature sensor is disposed on the contact surface area of ​​the smart wearable device, and the temperature sensor is used to collect body surface temperature signals.

[0014] The technical solution of this invention forms a four-level serial verification chain by sequentially performing wearing posture verification, temperature verification, human bioelectrical verification, and photoelectric signal verification. Only when all four levels of verification are passed is the device determined to be in a normal wearing state. This effectively eliminates the possibility of misjudgment by a single sensor and improves the accuracy of wearing detection. At the same time, by setting up four different types of sensors—posture sensor, temperature sensor, electrical sensor, and optical sensor—to detect from four different physical dimensions—device posture, body surface temperature, human bioelectrical signals, and blood flow signals—abnormalities or interference in a single dimension will not directly lead to errors in the final judgment result, thus improving the reliability of the detection results. Furthermore, by establishing the execution of temperature verification, bioelectrical verification, and photoelectric signal verification on the basis of the previous level of verification, the simultaneous operation of the four sensors is avoided, which helps to reduce device power consumption. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of a structure of an embodiment of the smart wearable device wearing detection method provided by the present invention; Figure 2 A schematic diagram of another embodiment of the smart wearable device wearing detection method provided by the present invention; Figure 3 A schematic diagram of another embodiment of the smart wearable device wearing detection method provided by the present invention; Figure 4 A schematic diagram of another embodiment of the smart wearable device wearing detection method provided by the present invention; Figure 5 A schematic diagram of another embodiment of the smart wearable device wearing detection method provided by the present invention; Figure 6 This is a schematic diagram of another embodiment of the smart wearable device wearing detection method provided by the present invention.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] Currently, wear detection for smart wearable devices mainly relies on photoplethysmography (PPG), which involves emitting light signals and receiving reflected light from the skin, then judging the wearing status based on changes in blood flow. However, this method has many problems in practical use: reflected light signals are easily misinterpreted as being worn when the device is placed on a non-human surface such as a table; spurious signals caused by clothing can interfere with the detection results; vasoconstriction in low temperatures weakens blood flow signals, affecting the accuracy of the judgment; artifacts caused by muscle vibration during exercise can also easily lead to misjudgments; and it is difficult to distinguish between different scenarios such as close to the skin, close to a warm non-living surface, and static placement. These factors result in a high misjudgment rate for wear detection, easily distorted health monitoring data, and potential accidental triggering of device functions. Therefore, improving the accuracy of wear detection for smart wearable devices has become an urgent problem to be solved in this field.

[0020] To address this issue, the present invention proposes a method for detecting wearability in smart wearable devices, aiming to solve the problem of high false detection rates in existing smart wearable devices due to their reliance on a single PPG sensor.

[0021] Please see Figure 1 In one embodiment of the present invention, the smart wearable device wearing detection method is applied to a smart wearable device, which is equipped with an attitude sensor, an optical sensor, a temperature sensor, and an electrical sensor. The method includes: Step S100: The posture sensor collects the posture data of the smart wearable device and verifies the wearing posture based on the device posture data. Step S200: If the wearing posture verification is passed, the temperature sensor collects the body surface temperature signal and performs temperature verification based on the body surface temperature signal. Step S300: If the temperature verification is passed, the electrical sensor collects the human skin current signal and performs human bioelectric verification based on the skin current signal. Step S400: If the bioelectric verification is successful, the optical sensor collects blood flow signals and performs photoelectric signal verification based on the blood flow signals. If all verifications—including wearing posture, temperature, bioelectricity, and photoelectric signal—pass, the device is considered to be in a normal wearing state.

[0022] Optionally, in step S100, the attitude sensor collects attitude data from the smart wearable device, verifies the wearing attitude based on the device's attitude data, and obtains the wearing attitude verification result. The attitude sensor includes a three-axis accelerometer and a three-axis gyroscope, collecting acceleration data and angular velocity data. The acceleration data and angular velocity data are filtered to extract the mean acceleration, variance acceleration, zero-crossing rate of acceleration, root mean square of angular velocity, and the pitch and roll angles of the smart wearable device as attitude features. The mean acceleration is the average value of the synthesized acceleration vector over a period of time. The variance acceleration is the degree of dispersion of the acceleration data relative to the mean. The zero-crossing rate of acceleration is the number of times the acceleration waveform crosses the zero level per unit time. The root mean square of angular velocity is the square root of the average of the squares of the three-axis angular velocity components. The pitch and roll angles are obtained by attitude fusion calculation of the acceleration data and angular velocity data. The above posture characteristics are compared with preset wearing posture thresholds. If the average acceleration is within the range of 0.8g to 1.2g (g is the acceleration due to gravity), the pitch and roll angles are both within ±15 degrees, and the root mean square angular velocity is less than 10 degrees / second, then the wearing posture verification is deemed successful. If any of the above conditions are not met, then the wearing posture verification is deemed unsuccessful. In step S100, the average acceleration is used to measure the magnitude of the resultant acceleration on the device to determine whether the smart wearable device is in a normal gravity environment; the acceleration variance is used to measure the amplitude of the vibration of the smart wearable device; the acceleration zero-crossing rate is used to measure the frequency characteristics of the vibration of the smart wearable device; the root mean square angular velocity is used to measure the intensity of the rotational motion of the smart wearable device; and the pitch and roll angles are used to measure the tilt angle of the smart wearable device relative to the horizontal reference plane.

[0023] The aforementioned posture characteristics describe the spatial motion state of the device from multiple dimensions, and an anomaly in a single dimension will not lead to misjudgment of the posture verification results. When the smart wearable device is in a wearing posture, it rests naturally with the user's arm, its average acceleration is close to the acceleration due to gravity, and its pitch and roll angles are within the angle range corresponding to the natural posture of the user's arm, with a small root mean square angular velocity. When the smart wearable device is in a non-wearing posture, at least one of the above posture characteristics deviates from the preset threshold range. Therefore, by judging the combination of the above posture characteristics, it is possible to distinguish between non-wearing postures such as the smart wearable device lying flat on a table, the smart wearable device in free fall, and the smart wearable device being rapidly flipped by hand, and normal wearing postures. Only when all the above posture characteristics meet the preset threshold requirements will the next level of temperature verification be performed, thereby improving the problem of activating subsequent high-power sensors when the wearing posture is obviously not met, and reducing the power consumption of the smart wearable device.

[0024] Optionally, if the wearing posture verification passes, step S200 is executed. The temperature sensor is activated and collects the body surface temperature signal, performs temperature verification based on the body surface temperature signal, and obtains the temperature verification result. The temperature sensor is deployed on the surface of the smart wearable device that contacts the skin, converts the temperature signal into an electrical signal output, and generates digital temperature data after analog-to-digital conversion. The temperature data is processed by moving average filtering to output a stable temperature value. Moving average filtering is performed by arithmetically averaging the temperature values ​​of a preset number of sampling points before and after the current sampling point to suppress instantaneous fluctuations in the temperature signal. The temperature value is compared with a preset human body surface temperature threshold range. If the temperature value is within the range of 32℃ to 37.5℃, the temperature verification is considered passed. If the temperature value is below 32℃ or above 37.5℃, the temperature verification is considered failed. The temperature range of the human body surface under normal physiological conditions is usually between 32℃ and 37.5℃, which is determined by the balance between heat production from human metabolism and heat dissipation from the body surface. When a smart wearable device is in contact with human skin, the temperature sensed by the temperature sensor falls within this range. When the smart wearable device is not in contact with human skin, such as when the device is placed on a table, clothing surface, or other non-human surface, the temperature sensed by the temperature sensor is usually below 32°C, or above 37.5°C due to contact with external heat sources (such as hand warmers or hot water bottles). In both cases, the temperature verification fails. When the smart wearable device is in contact with human skin but separated by clothing, the insulation effect of the fabric layer causes the temperature sensed by the temperature sensor to be below 32°C, also failing the temperature verification. Therefore, by comparing the temperature value with the threshold range of 32°C to 37.5°C, it is possible to distinguish between cases where the contact surface is human skin and cases where the contact surface is a non-human surface, and to differentiate between cases where the smart wearable device is directly in contact with human skin and cases where it is indirectly in contact (separated by clothing). When the temperature verification fails, the subsequent detection process is terminated, and the optical and electrical sensors are not activated, thereby improving the problem of performing subsequent bioelectrical and photoelectric signal verification in non-human surface contact scenarios and reducing the power consumption of the smart wearable device.

[0025] Optionally, if the temperature verification passes, step S300 is executed. The electrical sensor is activated and collects human skin current signals. Based on these skin current signals, human bioelectric verification is performed to obtain the bioelectric verification result. The electrical sensor includes two ECG electrode pads, symmetrically deployed on both sides of the contact surface of the smart wearable device. The ECG electrode pads output excitation signals to the skin and detect the response current flowing through the skin. The skin conductance value is calculated based on the excitation signal and the response current. The ECG electrode pads simultaneously collect the potential difference signal between two points on the skin surface. This potential difference signal contains an electrocardiogram (ECG) component and a noise component. The potential difference signal is converted to the frequency domain using a Fourier transform. The energy in the frequency band containing the ECG component is extracted as the effective signal strength, and the out-of-band energy is extracted as the noise intensity. The ratio of the effective signal strength to the noise intensity is calculated to obtain the signal-to-noise ratio (SNR) of the ECG signal. The calculated value is compared with a preset bioelectric threshold. If the skin conductance value is within the range of 10 microsiemens to 100 microsiemens and the SNR of the ECG signal is greater than 15 dB, the bioelectric verification is considered successful. If the skin conductivity is below 10 microsiemens or above 100 microsiemens, or the signal-to-noise ratio (SNR) of the ECG signal is less than or equal to 15 dB, the bioelectrical verification is considered unsuccessful. Human skin possesses a certain degree of conductivity due to its water and electrolyte content, with a conductivity typically ranging from 10 to 100 microsiemens. Non-living surfaces (such as desktops, plastics, and metals) or insulating surfaces do not possess this conductivity, and their conductivity is below 10 microsiemens. The periodic electrical activity of the human heart generates an ECG signal, which is conducted through tissues to the body surface and can be detected by ECG electrodes. The effective ECG signal typically has a SNR greater than 15 dB. Non-living surfaces do not generate ECG signals; the detected signal is primarily environmental noise, with a SNR below 15 dB. Therefore, by combining skin conductivity and ECG SNR, it is possible to distinguish living human skin from non-living surfaces. When bioelectric verification fails, the subsequent detection process is terminated and the optical sensor is not activated, thereby improving the problem of performing subsequent photoelectric signal verification in non-living surface contact scenarios and reducing the power consumption of smart wearable devices.

[0026] Optionally, the optical sensor is activated and acquires blood flow signals. Based on these signals, photoelectric signal verification is performed to obtain the verification result. Specifically, the optical sensor is a PPG sensor deployed on the skin-contacting surface of the smart wearable device, comprising a combination of a green LED and a photodiode. The green LED emits a wavelength of 525 nanometers and has a detection distance of 0.1 to 0.3 millimeters. The green LED emits light signals towards the skin, penetrating the skin surface to illuminate subcutaneous capillaries. Hemoglobin in the blood absorbs green light. During cardiac contraction, the blood volume in the blood vessels increases, the hemoglobin content increases, the absorption of green light increases, and the intensity of reflected light decreases. During cardiac relaxation, the blood volume in the blood vessels decreases, the hemoglobin content decreases, the absorption of green light decreases, and the intensity of reflected light increases. The photodiode receives the light signal reflected from the skin and converts it into an electrical signal for output. The intensity of this electrical signal fluctuates periodically with the periodic changes in blood volume caused by cardiac contraction and relaxation, forming a pulse wave signal. Because light signals are affected by subcutaneous tissue, bone, and blood when penetrating the skin, with blood absorbing green light much more readily than surrounding tissues, the electrical signal output by the photodiode carries the pulse wave characteristics of blood volume. The peak pulse wave value and signal-to-noise ratio (SNR) are extracted from the pulse wave signal. The peak pulse wave value is the maximum reflected light intensity within a single cardiac cycle, reflecting the optical signal intensity when the blood volume in the blood vessels reaches its peak during systole, and is positively correlated with the amplitude of blood volume change. The SNR is obtained through frequency domain analysis of the pulse wave signal. The pulse wave signal is converted to the frequency domain using a Fourier transform, and the energy within the 0.5Hz to 10Hz frequency band is extracted as the effective signal intensity. This frequency band covers the normal human heart rate range. The out-of-band energy is used as the noise intensity. The ratio of the effective signal intensity to the noise intensity is calculated, and the logarithm to base 10 is multiplied by 20 to obtain the SNR in dB. The extracted pulse wave peak value is compared with a first preset benchmark value, which is calibrated based on pulse wave peak values ​​collected under standard test conditions before the device leaves the factory. The extracted signal-to-noise ratio (SNR) is compared with a preset SNR threshold. When the pulse wave peak value is greater than 60% of the first preset benchmark value and the SNR is greater than 12dB, the photoelectric signal verification is considered successful. If the pulse wave peak value is less than or equal to 60% of the first preset benchmark value, it indicates that the intensity fluctuation of the reflected light signal is insufficient, i.e., the change in blood volume in subcutaneous blood vessels is lower than the normal level under wearing conditions. The possible reasons are that the smart wearable device is not in close contact with the skin or the contact surface is non-human living tissue. If the SNR is less than or equal to 12dB, it indicates that the ratio of the effective pulse wave signal intensity to the background noise intensity is insufficient, i.e., the periodic blood flow component in the reflected light signal is submerged by noise. The possible reasons are poor device contact, ambient light interference, or the contact surface not having periodic blood flow characteristics.When both the pulse wave peak value and the signal-to-noise ratio simultaneously meet preset conditions, it indicates that there is a periodic blood flow change driven by the heartbeat beneath the contact surface of the smart wearable device, and the intensity of this blood flow signal is consistent with the characteristics of a normal human pulse. If the bioelectric verification or photoelectric signal verification fails, the subsequent detection process is terminated or the device is determined to be not being worn.

[0027] When the posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification all pass, the smart wearable device is determined to be in a normal wearing state, and a wearing confirmation signal is output. If any of the above four levels of verification fails, the smart wearable device is determined to be in a non-wearing or invalid wearing state, and no wearing confirmation signal is output. In steps S100 to S400, the posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification are executed sequentially in a fixed order. Each subsequent verification is only initiated after the previous verification is passed. If any verification fails, the subsequent verification process is terminated, and the device is determined to be in a non-wearing state. In this sequential execution method, the posture sensor remains active, the temperature sensor is activated only after the posture verification is passed, the electrical sensor is activated only after the temperature verification is passed, and the optical sensor is activated only after the bioelectrical verification is passed. The temperature sensor, electrical sensor, and optical sensor do not activate if their respective preceding verification fails. Due to the differences in power consumption of different sensors, the optical sensor has higher power consumption because it needs to continuously drive the light-emitting element, followed by the electrical sensor, then the temperature sensor, and finally the posture sensor. The aforementioned step-by-step activation method reduces the operating time of high-power sensors and lowers the overall power consumption of smart wearable devices. Simultaneously, posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification verify the wearing status from mechanical, thermal, electrical, and optical dimensions, respectively. These four dimensions are independent yet complementary; only when all four dimensions pass verification is the device considered to be in a normal wearing state, avoiding misjudgments caused by environmental interference or physiological changes from a single sensor.

[0028] The technical solution of this invention forms a four-level serial verification chain by sequentially performing wearing posture verification, temperature verification, human bioelectrical verification, and photoelectric signal verification. Only when all four levels of verification are passed is the device determined to be in a normal wearing state. This effectively eliminates the possibility of misjudgment by a single sensor and improves the accuracy of wearing detection. At the same time, by setting up four different types of sensors—posture sensor, temperature sensor, electrical sensor, and optical sensor—to detect from four different physical dimensions—device posture, body surface temperature, human bioelectrical signals, and blood flow signals—abnormalities or interference in a single dimension will not directly lead to errors in the final judgment result, thus improving the reliability of the detection results. Furthermore, by establishing the execution of temperature verification, bioelectrical verification, and photoelectric signal verification on the basis of the previous level of verification, the simultaneous operation of the four sensors is avoided, which helps to reduce device power consumption.

[0029] like Figure 2 As shown, in one embodiment, the posture sensor collects posture data of the smart wearable device, and the step of verifying the wearing posture based on the device posture data includes: Step S110: The attitude sensor acquires acceleration and angular velocity data; Step S120: Filter the acceleration and angular velocity data; Step S130: Based on the filtered acceleration data and angular velocity data, extract the mean acceleration, variance acceleration, rate of acceleration crossing zero, root mean square angular velocity, pitch angle and roll angle of the smart wearable device as attitude features. Step S140: When the average acceleration is 0.8g~1.2g, the pitch and roll angles are both within ±15°, and the root mean square angular velocity is less than 10°. o If the wearing posture verification is successful at a rate of / s, the wearing posture verification is considered successful.

[0030] Optionally, in step S110, the attitude sensor acquires acceleration and angular velocity data. The attitude sensor includes a three-axis accelerometer and a three-axis gyroscope. The three-axis accelerometer detects the acceleration components of the smart wearable device along three orthogonal axes, and the three-axis gyroscope detects the angular velocity components of the smart wearable device along three orthogonal axes. The acceleration data includes X-axis acceleration values, Y-axis acceleration values, and Z-axis acceleration values, and the angular velocity data includes X-axis angular velocity values, Y-axis angular velocity values, and Z-axis angular velocity values. The sampling frequency is set to 100Hz, the acquisition time is 1 second, and 100 sets of acceleration sampling data and 100 sets of angular velocity sampling data are obtained.

[0031] Optionally, in step S120, the acceleration and angular velocity data are filtered. The filtering process employs a combination of a low-pass filter and a Kalman filter. The low-pass filter attenuates high-frequency noise components in the acceleration and angular velocity data, retaining the effective low-frequency signal components. The Kalman filter, based on the state estimate from the previous moment and the observed value at the current moment, performs an optimal estimate of the state value at the current moment, removing environmental vibration noise and data offset introduced by sensor zero-bias drift.

[0032] Optionally, in step S130, based on the filtered acceleration and angular velocity data, the mean acceleration, variance acceleration, zero-crossing rate of acceleration, root mean square of angular velocity, and the pitch and roll angles of the smart wearable device are extracted as attitude features. The mean acceleration is the average value of the magnitude of the three-axis acceleration composite vector over the acquisition period. It is calculated by taking the square root of the sum of the squares of the X-axis, Y-axis, and Z-axis acceleration values ​​in each of the 100 sets of sampled data to obtain the composite acceleration magnitude at that moment, and then arithmetically averaging the 100 composite acceleration magnitudes. The variance acceleration is the average of the squared deviations of the composite acceleration magnitude from the mean acceleration at each moment. The zero-crossing rate of acceleration is the number of times the difference between the composite acceleration magnitude and the mean acceleration intersects with zero level within the acquisition period. The root mean square (RMS) of angular velocity is the root mean square value of the composite vector of the three-axis angular velocities. It is calculated by taking the square root of the sum of the squares of the X-axis, Y-axis, and Z-axis angular velocities in each of the 100 sets of sampled data to obtain the composite angular velocity magnitude at that moment. Then, the sum of the squares of the 100 composite angular velocity magnitudes is averaged, and the square root is taken. Pitch and roll angles are obtained by attitude fusion calculation using filtered acceleration and angular velocity data. Attitude fusion calculation employs the quaternion method, taking acceleration and angular velocity data as input, updating the quaternions through quaternion differential equations, converting the quaternions to Euler angles, and extracting the pitch and roll angles from the Euler angles.

[0033] Optionally, in step S140, the above-mentioned posture characteristics are compared with preset wearing posture thresholds. If the average acceleration is within the range of 0.8g to 1.2g, the pitch and roll angles are both within ±15 degrees, and the root mean square angular velocity is less than 10 degrees / second, then the wearing posture verification is deemed successful. If any of the above conditions are not met, then the wearing posture verification is deemed unsuccessful.

[0034] like Figure 3 As shown, in one embodiment, the temperature sensor collects a body surface temperature signal, and the step of verifying the temperature based on the body surface temperature signal includes: Step S210: The temperature sensor collects body surface temperature data; Step S220: Perform moving average filtering on the body surface temperature data and output the temperature value; Step S230: If the temperature value is within the range of 32℃~37.5℃, the temperature verification is deemed successful.

[0035] Optionally, in step S210, the temperature sensor collects body surface temperature data. The temperature sensor is deployed in the center area of ​​the surface of the smart wearable device that contacts the skin, and uses a high-precision NTC thermistor or infrared temperature sensor. The temperature sensor is in direct contact with the skin, converting the temperature signal into an electrical signal output, which is then converted from analog to digital to generate digital temperature data. The temperature sensor has a sampling accuracy of ±0.1℃, a sampling frequency of 50Hz, and a collection time of 0.5 seconds, obtaining 25 sets of temperature sampling data.

[0036] Optionally, in step S220, the body surface temperature data is subjected to a moving average filtering process, and the temperature value is output. The moving average filtering suppresses instantaneous fluctuations in the temperature signal by performing an arithmetic average of the temperature values ​​of a preset number of sampling points before and after the current sampling point. Specifically, the sliding window width is set to 5, meaning that the arithmetic average of the temperature values ​​of the current sampling point and the two sampling points before and after it (a total of 5 points) is performed to obtain the filtered temperature value at that moment. This moving average filtering process is sequentially performed on 25 sets of temperature sampling data to output stable temperature values.

[0037] Optionally, in step S230, the temperature value is compared with a preset human body surface temperature threshold range. If the temperature value is within the range of 32℃ to 37.5℃, the temperature verification is deemed successful. If the temperature value is lower than 32℃ or higher than 37.5℃, the temperature verification is deemed unsuccessful.

[0038] In the above embodiments, the temperature sensor is awakened and starts data acquisition after the wearing posture verification is passed, and remains in a dormant state when the wearing posture verification fails. The moving average filtering process can suppress instantaneous fluctuations in the temperature signal, outputting a stable temperature value and avoiding unstable verification results caused by temperature signal jumps. The temperature range of the human body surface under normal physiological conditions is typically between 32°C and 37.5°C, determined by the balance between heat production from human metabolism and heat dissipation from the body surface. When the smart wearable device is in contact with human skin, the temperature sensed by the temperature sensor falls within this range. When the smart wearable device is not in contact with human skin, such as when it is placed on a table, clothing surface, or other non-human surface, the temperature sensed by the temperature sensor is typically below 32°C. When the device is in contact with an external heat source, such as a hand warmer or hot water bottle, the temperature sensed by the temperature sensor may be above 37.5°C. When the smart wearable device is in contact with human skin but separated by clothing, the insulation effect of the fabric layer causes the temperature sensed by the temperature sensor to be below 32°C. Temperature values ​​below 32°C or above 37.5°C will not pass the temperature verification. Therefore, by comparing the temperature value with a threshold range of 32℃ to 37.5℃, it is possible to distinguish between cases where the contact surface is human skin and cases where the contact surface is a non-human surface, and to differentiate between direct and indirect contact (through clothing) between the device and human skin. If the temperature verification fails, the subsequent testing process is terminated, and the electrical and optical sensors are not activated, thus avoiding subsequent bioelectrical and photoelectric signal verifications in non-human surface contact scenarios and reducing device power consumption. Temperature verification is only performed after posture verification is passed, and the temperature sensor only operates during temperature verification, remaining in a dormant state at other times, further reducing device power consumption.

[0039] like Figure 4 As shown, in one embodiment, an electrical sensor collects human skin current signals, and the step of verifying human bioelectricity based on the skin current signals includes: Step S310: The electrical sensor collects skin current signals and electrocardiogram signals; Step S320: Calculate the signal-to-noise ratio of the skin current signal and the electrocardiogram signal; Step S330: When the skin conductance value is between 10μS and 100μS and the signal-to-noise ratio of the electrocardiogram signal is greater than 15dB, the bioelectric verification is deemed to be successful.

[0040] Optionally, in step S310, the electrical sensor acquires skin current signals and electrocardiogram (ECG) signals. The electrical sensor includes two ECG electrode pads made of silver-silver chloride conductive material, symmetrically deployed on both sides of the contact surface of the smart wearable device. The electrode pads are made of silver-silver chloride conductive material, with an electrode spacing of 20 to 30 mm and an input impedance greater than or equal to 10 MΩ. After the ECG electrode pads contact the skin, they output an excitation signal to the skin and detect the response current flowing through the skin to obtain the skin current signal. Simultaneously, the ECG electrode pads acquire the potential difference signal between two points on the skin surface. This potential difference signal includes an ECG component and a noise component to obtain the ECG signal. The acquisition time is 2 seconds, and the sampling frequency is set to 250 Hz.

[0041] Optionally, in step S320, the signal-to-noise ratio (SNR) of the skin conductance and the ECG signal is calculated. The skin conductance is calculated by dividing the voltage of the excitation signal applied to the skin by the detected response current value, in micro-Siemens. The SNR of the ECG signal is calculated as follows: the acquired time-domain ECG signal is transformed to the frequency domain using a Fourier transform to obtain the spectral distribution of the ECG signal. The energy within the 0.5Hz to 40Hz frequency band is extracted as the effective ECG signal intensity, and the out-of-band energy is extracted as the noise intensity. The ratio of the effective ECG signal intensity to the noise intensity is calculated, and the logarithm to base 10 is multiplied by 20 to obtain the SNR in dB.

[0042] Optionally, in step S330, the calculated value is compared with a preset bioelectric threshold. If the skin conductance value is within the range of 10 microsiemens to 100 microsiemens and the signal-to-noise ratio of the electrocardiogram signal is greater than 15 dB, the bioelectric verification is deemed successful. If the skin conductance value is lower than 10 microsiemens or higher than 100 microsiemens, or the signal-to-noise ratio of the electrocardiogram signal is less than or equal to 15 dB, the bioelectric verification is deemed unsuccessful.

[0043] In the above embodiments, human skin possesses a certain degree of conductivity due to its content of water and electrolytes, with a conductivity value typically ranging from 10 microsiemens to 100 microsiemens. Non-living surfaces or insulating surfaces do not possess this conductivity characteristic, and their conductivity values ​​are below 10 microsiemens. The periodic electrical activity of the human heart generates an electrocardiogram (ECG) signal, which is conducted through tissue to the body surface and can be detected by ECG electrode pads. The signal-to-noise ratio (SNR) of an effective ECG signal is typically greater than 15 dB. Non-living surfaces do not generate ECG signals; the detected signal is primarily environmental noise, with a SNR below 15 dB. Therefore, by combining skin conductivity and ECG SNR, it is possible to distinguish living human skin from non-living surfaces. Bioelectric verification is only performed after temperature verification is passed. The electrical sensor only operates during bioelectric verification and remains dormant at other times, reducing device power consumption. If bioelectric verification fails, the subsequent detection process is terminated, and the optical sensor does not activate, thereby avoiding subsequent photoelectric signal verification in non-living surface contact scenarios and further reducing device power consumption.

[0044] like Figure 5 As shown, in one embodiment, the step of using an optical sensor to acquire blood flow signals and performing photoelectric signal verification based on the blood flow signals includes: Step S410: The optical sensor acquires blood flow signals; Step S420: Extract the peak pulse wave value and signal-to-noise ratio of the blood flow signal; Step S430: If the peak value of the pulse wave is greater than 60% of the first preset reference value and the signal-to-noise ratio is greater than 12dB, the photoelectric signal verification is determined to be successful.

[0045] Optionally, in step S410, an optical sensor acquires blood flow signals. Specifically, the optical sensor is a PPG sensor deployed on the skin-contacting surface of the smart wearable device, comprising a combination of a green LED and a photodiode. The green LED emits light signals at a wavelength of 525 nm, with a detection distance of 0.1 to 0.3 mm. The green LED emits light signals towards the skin, penetrating the skin surface to illuminate subcutaneous capillaries. The photodiode receives the light signals reflected back from the skin and converts them into electrical signals for output. The intensity of this electrical signal fluctuates periodically with the periodic changes in blood volume within the blood vessels caused by cardiac contraction and relaxation, forming a pulse wave signal. The acquisition duration is 3 seconds, and the sampling frequency is set to 125 Hz.

[0046] Optionally, in step S420, the pulse wave peak value and signal-to-noise ratio (SNR) are extracted from the pulse wave signal. The pulse wave peak value is the maximum intensity of reflected light within a single cardiac cycle. The SNR is obtained by performing frequency domain analysis on the pulse wave signal. The pulse wave signal is transformed to the frequency domain using a Fourier transform, and the energy within the 0.5Hz to 10Hz frequency band is extracted as the effective signal intensity. This frequency band covers the normal heart rate range of the human body. The out-of-band energy is used as the noise intensity. The ratio of the effective signal intensity to the noise intensity is calculated, and the logarithm to base 10 is multiplied by 20 to obtain the SNR in dB.

[0047] Optionally, in step S430, the extracted pulse wave peak value is compared with a first preset reference value, and the extracted signal-to-noise ratio (SNR) is compared with a preset SNR threshold. The first preset reference value is obtained based on the pulse wave peak value collected under standard test conditions before the device leaves the factory. If the pulse wave peak value is greater than 60% of the first preset reference value and the SNR is greater than 12dB, the photoelectric signal verification is deemed successful. If the pulse wave peak value is less than or equal to 60% of the first preset reference value, or the SNR is less than or equal to 12dB, the photoelectric signal verification is deemed unsuccessful.

[0048] In the above embodiments, the blood volume in the subcutaneous blood vessels changes regularly with the periodic beating of the heart, causing the intensity of reflected light to fluctuate periodically in sync with the heartbeat. The pulse wave signal can reflect the presence of periodic blood flow driven by the heartbeat in the subcutaneous tissue. When the smart wearable device is attached to the skin, the optical sensor can detect this periodic blood flow signal, which is manifested as a pulse wave peak reaching a certain amplitude and a high signal-to-noise ratio. A pulse wave peak greater than 60% of a first preset reference value indicates that the intensity fluctuation amplitude of the reflected light signal reaches the normal level under wearing conditions, meaning the change in blood volume in the subcutaneous blood vessels is consistent with the characteristics of a normal human pulse. A signal-to-noise ratio greater than 12dB indicates that the ratio of the effective pulse wave signal intensity to the background noise intensity is sufficient, meaning the periodic blood flow component in the reflected light signal is significantly stronger than the background noise, and the signal quality meets the judgment requirements. When the device is attached to a non-living surface, there is no periodic blood flow change, the signal detected by the optical sensor does not contain a periodic pulse wave component, the pulse wave peak is less than 60% of the first preset reference value, or the signal-to-noise ratio is less than 12dB. When the device does not fit snugly against the skin, the periodic blood flow component in the reflected light signal weakens, resulting in insufficient pulse wave peak value or signal-to-noise ratio. Therefore, by combining the pulse wave peak value and signal-to-noise ratio, it is possible to ultimately confirm the presence of periodic blood flow changes driven by the heartbeat beneath the contact surface of the smart wearable device. Photoelectric signal verification is only performed after bioelectric verification is successful. The optical sensor only operates during photoelectric signal verification, remaining in a dormant state at other times. Optical sensors consume significant power due to the need to continuously drive the light-emitting element; this operating mode reduces the operating time of the optical sensor, thereby lowering the device's power consumption.

[0049] like Figure 6 As shown, in one embodiment, the smart wearable device wearing detection method further includes a re-detection step; If any of the verifications—wearing posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification—fail, record the failed verification dimension and the corresponding sensor data. Initiate a delayed re-examination mechanism, re-execute the verification of failed dimensions after a 2-second interval, and adopt a dynamic threshold adjustment strategy during the re-examination process to adjust the current verification threshold to a threshold that is suitable for the current detection scenario; If the threshold is adjusted and the verification is successful, proceed with the subsequent verification steps. If the test fails three times, the device is deemed to be not worn or worn in an invalid state.

[0050] Optionally, if any of the following verifications fails—wearing posture verification, temperature verification, bioelectrical conductivity verification, and photoelectric signal verification—the failed verification dimension and the corresponding sensor data are recorded. Specifically, when the wearing posture verification fails, the posture verification dimension and the corresponding acceleration and angular velocity data are recorded; when the temperature verification fails, the temperature verification dimension and the corresponding temperature data are recorded; when the bioelectrical conductivity verification fails, the bioelectrical conductivity verification dimension and the corresponding skin current signal data and electrocardiogram signal data are recorded; when the photoelectric signal verification fails, the photoelectric signal verification dimension and the corresponding pulse wave signal data are recorded. The recorded data includes the specific time of the failed verification, the failed threshold item (e.g., average acceleration value exceeding the range, temperature value exceeding the range, skin conductance value exceeding the range, or insufficient electrocardiogram signal-to-noise ratio, insufficient pulse wave peak value, or insufficient signal-to-noise ratio), and the original sampling data at the time of failure.

[0051] Optionally, a delayed re-examination mechanism can be activated, re-executing the failed dimension verification after a 2-second interval. During the re-examination process, a dynamic threshold adjustment strategy is employed to adjust the current verification threshold to a threshold suitable for the current detection scenario. Specifically, when a certain level of verification fails, a 2-second wait is performed before re-executing the complete process for that level of verification. Taking posture verification as an example, the process of collecting acceleration and angular velocity data from the posture sensor, filtering, extracting posture features, and comparing the posture features with a preset threshold is re-executed. Taking temperature verification as an example, the process of collecting body surface temperature data from the temperature sensor, performing moving average filtering, and comparing the output temperature value with a preset threshold is re-executed. Taking bioelectric verification as an example, the process of collecting skin current and electrocardiogram signals from the electrical sensor, calculating skin conductance and electrocardiogram signal-to-noise ratio, and comparing the calculation results with a preset threshold is re-executed. Taking photoelectric signal verification as an example, the process of collecting blood flow signals from the optical sensor, extracting pulse wave peak value and signal-to-noise ratio, and comparing the extraction results with a preset threshold is re-executed. During the re-examination process, the current verification threshold is adjusted to a threshold suitable for the current detection scenario. The current detection scenario is identified using the root mean square (RMS) angular velocity detected by the attitude sensor, the temperature value detected by the temperature sensor, and the ambient temperature value detected by the ambient temperature sensor. When the RMS angular velocity is greater than or equal to 10 degrees / second, it is identified as a motion scenario, and the angular velocity threshold of the attitude sensor is adjusted from 10 degrees / second to 15 degrees / second. When the ambient temperature is below 10°C, it is identified as a low-temperature scenario, and the lower limit of the temperature verification threshold is adjusted from 32°C to 30°C, and the lower limit of the skin conductance threshold for bioelectricity verification is adjusted from 10 micro-Siemens to 8 micro-Siemens. When the ambient temperature is above 35°C, it is identified as a high-temperature scenario, and the upper limit of the temperature verification threshold is adjusted from 37.5°C to 38°C. Except for the above adjustments, the other verification thresholds remain unchanged. After adjusting the thresholds, the verification at this level is re-executed. If the verification passes, the subsequent verification steps continue; if the verification fails, the verification is re-executed after waiting for 2 seconds, with a maximum of 3 re-tests. If the thresholds are adjusted and the verification passes, the subsequent verification steps are executed. If the test fails three times, the device is deemed to be not worn or worn in an invalid state.

[0052] In the above implementation, when a certain level of verification fails, the failed verification dimension and corresponding sensor data are recorded for subsequent analysis and debugging, helping to determine the specific reason for the verification failure. A 2-second delay before retesting avoids misjudgments caused by momentary interference, such as a user briefly raising their hand, sudden fluctuations in ambient temperature, or momentary poor contact between the electrode pad and skin, allowing time for the detection conditions to stabilize. The dynamic threshold adjustment strategy adapts the verification threshold during retesting to the current detection scenario. In motion scenarios, the angular velocity of the human arm naturally increases; in low-temperature environments, the human body surface temperature naturally decreases, and the skin conductivity naturally decreases; in high-temperature environments, the human body surface temperature naturally increases. These changes are all within the normal physiological or behavioral range and do not indicate that the smart wearable device is not being worn. Through dynamic threshold adjustment, misjudgments of not wearing due to overly strict threshold settings in the above scenarios can be eliminated, reducing misjudgments caused by scene changes while maintaining detection accuracy. Simultaneously, the number of retests is limited to 3 to avoid increased power consumption and response delays caused by unlimited retesting, achieving a balance between improving the success rate of wearing status determination and maintaining low power consumption. If the device fails the retest three times, it is determined to be in a state of not wearing or invalid wearing, and the subsequent verification process is terminated. The smart wearable device will remain in or enter a low power mode and will not trigger the health monitoring function or notification reminder function.

[0053] The present invention also proposes a control device, which includes a memory, a processor, and a control program for detecting the wearing of a smart wearable device stored in the memory and executable on the processor. The control program for detecting the wearing of a smart wearable device is configured to implement the smart wearable device detection method as described above.

[0054] Specifically, the control device is located inside the smart wearable device and is electrically connected to the posture sensor, temperature sensor, electrical sensor, and optical sensor. The control device can be implemented using a Cortex-M4 core MCU (such as the STM32L476), which integrates a Sensor Hub module to support simultaneous acquisition and processing of data from multiple sensors. The memory includes non-volatile storage media (such as Flash memory) and volatile storage media (such as SRAM). The non-volatile storage media is used to store the control program for the smart wearable device's wear detection and various preset verification thresholds (including posture verification thresholds, temperature verification thresholds, bioelectrical verification thresholds, photoelectric signal verification thresholds, and scene thresholds corresponding to dynamic threshold adjustment strategies). The volatile storage media is used to temporarily store the raw data collected by the sensors during program execution (including acceleration data, angular velocity data, temperature data, skin current signal data, electrocardiogram signal data, and pulse wave signal data) and intermediate calculated data (including filtered data, extracted posture feature values, temperature values, skin conductance values, electrocardiogram signal-to-noise ratio, pulse wave peak value, and signal-to-noise ratio).

[0055] It is worth noting that since the control device of the present invention includes the above-mentioned smart wearable device wearing detection method, the control device of the present invention also includes all the above-mentioned smart wearable device wearing detection methods and the effects of each of the embodiments, which will not be repeated here.

[0056] The present invention also proposes a smart wearable device, which includes an attitude sensor, an optical sensor, a temperature sensor, an electrical sensor, and the control device described above.

[0057] It is worth noting that since the smart wearable device of the present invention includes the above-mentioned control device, the smart wearable device of the present invention also includes embodiments of all the above-mentioned control devices and the effects brought by each embodiment, which will not be repeated here.

[0058] In one embodiment, an attitude sensor is disposed inside the smart wearable device, and the attitude sensor is used to collect acceleration data and angular velocity data; An optical sensor is placed on the surface of the smart wearable device that contacts the skin. The optical sensor is used to collect light signals reflected from the skin and output blood flow signals. Electrical sensors are symmetrically arranged on both sides of the contact surface of the smart wearable device. These electrical sensors are used to collect electrical signals from human skin. A temperature sensor is placed on the contact surface of the smart wearable device to collect body surface temperature signals.

[0059] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for detecting the wearing status of a smart wearable device, applied to a smart wearable device, characterized in that, The smart wearable device is equipped with an attitude sensor, an optical sensor, a temperature sensor, and an electrical sensor. The method includes: The posture sensor collects posture data of the smart wearable device and verifies the wearing posture based on the device posture data. If the wearing posture verification is passed, the temperature sensor collects the body surface temperature signal and performs temperature verification based on the body surface temperature signal. If the temperature verification is passed, the electrical sensor collects human skin current signals and performs human bioelectric verification based on the skin current signals. If the bioelectric verification is successful, the optical sensor acquires blood flow signals and performs photoelectric signal verification based on the blood flow signals. If the wearing posture verification, temperature verification, bioelectrical verification, and photoelectric signal verification all pass, it is determined to be a normal wearing state.

2. The smart wearable device wearing detection method as described in claim 1, characterized in that, The posture sensor collects posture data from the smart wearable device, and the step of verifying the wearing posture based on the device posture data includes: The attitude sensor collects acceleration and angular velocity data; The acceleration data and the angular velocity data are filtered. Based on the filtered acceleration data and angular velocity data, the mean acceleration, variance acceleration, rate of acceleration crossing zero, root mean square angular velocity, pitch angle and roll angle of the smart wearable device are extracted as attitude features. The average acceleration is between 0.8g and 1.2g, the pitch angle and roll angle are both within ±15°, and the root mean square angular velocity is less than 10°. o If the wearing posture verification is successful, the result is determined to be / s.

3. The method for detecting the wearing of a smart wearable device as described in claim 1, characterized in that, The temperature sensor collects body surface temperature signals, and the step of verifying temperature based on the body surface temperature signals includes: The temperature sensor collects body surface temperature data; The body surface temperature data is processed by a moving average filter, and the temperature value is output. If the temperature value is within the range of 32℃ to 37.5℃, the temperature verification is deemed successful.

4. The method for detecting the wearing of a smart wearable device as described in claim 1, characterized in that, The steps for verifying human bioelectricity based on the human skin current signal acquired by the electrical sensor include: The electrical sensor collects skin current signals and electrocardiogram signals; Calculate the signal-to-noise ratio of the skin current signal and the electrocardiogram signal; If the skin conductance value is between 10 μS and 100 μS and the signal-to-noise ratio of the electrocardiogram signal is greater than 15 dB, the bioelectric verification is deemed to be successful.

5. The method for detecting the wearing of a smart wearable device as described in claim 1, characterized in that, The step of using the optical sensor to acquire blood flow signals and performing photoelectric signal verification based on the blood flow signals includes: The optical sensor collects blood flow signals; Extract the pulse wave peak value and signal-to-noise ratio of the blood flow signal; If the peak value of the pulse wave is greater than 60% of the first preset reference value and the signal-to-noise ratio is greater than 12dB, the photoelectric signal verification is deemed successful.

6. The method for detecting the wearing of a smart wearable device as described in claim 1, characterized in that, The smart wearable device wearing detection method also includes a re-detection step; If any one of the following verifications fails: wearing posture verification, temperature verification, bioelectric verification, and photoelectric signal verification, the failed verification dimension and the corresponding sensor data shall be recorded. Initiate a delayed re-examination mechanism, re-execute the verification of failed dimensions after a 2-second interval, and adopt a dynamic threshold adjustment strategy during the re-examination process to adjust the current verification threshold to a threshold that is suitable for the current detection scenario; If the threshold is adjusted and the verification is successful, proceed with the subsequent verification steps. If the test fails three times, the device is deemed to be not worn or worn in an invalid state.

7. The smart wearable device wearing detection method as described in claim 6, characterized in that, The dynamic threshold adjustment strategy includes: In motion scenarios, the angular velocity threshold of the attitude sensor is adjusted from 10° / s to 15° / s; In an environment where the ambient temperature is below 10°C, the lower limit of the temperature threshold for temperature verification is adjusted from 32°C to 30°C, and the lower limit of the skin conductance threshold for bioelectric verification is adjusted from 10μS to 8μS. In environments where the ambient temperature is above 35°C, the upper limit of the temperature threshold for temperature verification is adjusted from 37.5°C to 38°C.

8. A control device, characterized in that, The control device includes: a memory, a processor, and a control program for detecting the wearing of a smart wearable device stored in the memory and executable on the processor, wherein the control program for detecting the wearing of a smart wearable device is configured to implement the smart wearable device wearing detection method as described in any one of claims 1 to 7.

9. A smart wearable device, characterized in that, The smart wearable device includes an attitude sensor, an optical sensor, a temperature sensor, an electrical sensor, and a control device as described in claim 8.

10. The smart wearable device as described in claim 9, characterized in that, The attitude sensor is installed inside the smart wearable device and is used to collect acceleration data and angular velocity data. The optical sensor is disposed on the surface of the smart wearable device that contacts the skin. The optical sensor is used to collect light signals reflected from the skin and output blood flow signals. The electrical sensors are symmetrically arranged on both sides of the contact surface of the smart wearable device, and the electrical sensors are used to collect the current signal of human skin. The temperature sensor is disposed on the contact surface area of ​​the smart wearable device, and the temperature sensor is used to collect body surface temperature signals.

Images