Neck posture monitoring system based on self-driven sensing

Abstract

The invention provides a neck posture monitoring system based on self-driven sensing, and the system comprises a monitoring part which is used for being attached to a human body, and obtaining a neck motion signal based on a self-driven sensing principle; and the control part is used for generating a neck posture according to the neck movement signal acquired by the monitoring part and determining whether the neck posture is a bad posture. The monitoring part acquires the neck movement signal based on the self-driven sensing principle, so that the monitoring part has the advantages of light weight, small size and convenience in wearing while accurately acquiring the neck movement signal; meanwhile, due to the fact that the monitoring part is in linkage with the control part, relevant data of the monitoring part can be conveniently and rapidly obtained through the control part, the monitoring part is controlled, people can conveniently adjust the bad posture of the neck, and the problems that an existing neck posture monitoring device cannot provide convenient and comfortable wearing experience and is poor in posture monitoring precision are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence and wearable devices, and in particular to a neck posture monitoring system based on self-driven sensing. Background Art

[0002] Cervical spondylosis is a disease caused by degenerative pathological changes in the cervical spine. The main causes are long-term strain of the cervical spine, bone hyperplasia or disc herniation, ligament thickening and other factors that compress the cervical spinal cord, nerve roots, and vertebral arteries. With the rapid development of society and the popularization of electronic products, people spend more and more time working and entertaining with their heads down, causing cervical spondylosis to gradually become one of the common health problems in the world. Symptoms of cervical spondylosis include neck pain, stiffness, numbness and weakness in the limbs, dizziness, nausea, vomiting, etc. In severe cases, it can manifest as blurred vision, tachycardia and difficulty swallowing, which greatly affects people's normal life.

[0003] If the neck posture can be monitored and corrected in the early stage of cervical spondylosis, the symptoms can be effectively controlled, recurrence can be reduced, and the quality of life can be improved. Therefore, the researchers proposed a neck posture monitoring device based on posture detection. At present, most neck posture monitoring devices use acceleration sensors, gyroscopes, photoelectric detectors, etc. to achieve real-time monitoring of neck posture.

[0004] However, existing neck posture monitoring devices based on traditional sensing technology are often complex in design and expensive. In order to ensure the accuracy of the sensor, they are often made of silicon-based hard materials, which makes them large in size and less comfortable to wear. Existing neck posture monitoring devices based on photoelectric detection technology require additional light sources, which makes the equipment larger and less portable. In addition, its monitoring effect is heavily dependent on the quality of the light source, which is easily affected by the environment and has poor stability. Existing neck posture monitoring devices based on multi-axis acceleration sensing technology need to obtain posture information at multiple points, so sensors need to be installed around the neck and even on the forehead. Not only is it complicated to wear and increases the burden on the head, but the recognition accuracy is also low in some postures (such as the head-up posture). Summary of the invention

[0005] The purpose of the present invention is to provide a neck posture monitoring system based on self-driven sensing to solve the problem that the existing neck posture monitoring device cannot provide a convenient and comfortable wearing experience and has poor posture monitoring accuracy.

[0006] In order to solve the above technical problems, the present invention provides a neck posture monitoring system based on self-driven sensing, comprising: The monitoring part is used to be attached to the human body and obtain the neck movement signal based on the self-driven sensing principle; The control unit is used to generate the neck posture according to the neck movement signal collected by the monitoring unit, and confirm whether the neck posture is a bad posture.

[0007] Optionally, in the neck posture monitoring system based on self-driven sensing, the monitoring part includes a wearable fabric and a self-driven sensing unit arranged on the side of the wearable fabric that contacts the human body surface; the wearable fabric is used to be worn on the human neck; the self-driven sensing unit includes a plurality of friction electric self-driven sensors, which are used to attach to the human neck to obtain neck movement signals.

[0008] Optionally, in the neck posture monitoring system based on self-driven sensing, the triboelectric self-driven sensor has a flexible substrate, and the flexible substrate is fixed to the wearable fabric.

[0009] Optionally, in the self-driven sensing-based neck posture monitoring system, a plurality of the triboelectric self-driven sensors are arranged in an array, and each of the triboelectric self-driven sensors generates its own sub-motion electrical signal.

[0010] Optionally, in the self-driven sensing-based neck posture monitoring system, the monitoring unit further includes a flexible data acquisition unit, and the flexible data acquisition unit is used to transmit the neck movement signal acquired by the self-driven sensing unit to the control unit in a wireless communication manner.

[0011] Optionally, in the neck posture monitoring system based on self-driven sensing, the flexible data acquisition unit includes a microcontroller, a signal collector, a wireless communication device and a power manager; the microcontroller, the signal collector and the wireless communication device are all connected to the power manager to obtain the required voltage; the signal collector and the wireless communication device are both connected to the microcontroller to be controlled by the microcontroller; the signal collector is connected to the self-driven sensing unit to obtain the neck motion signal, and filter and amplify the neck motion signal; the wireless communication device is wirelessly connected to the control unit to transmit the filtered and amplified neck motion signal to the control unit by wireless communication.

[0012] Optionally, in the neck posture monitoring system based on self-driven sensing, the signal collector includes multiple signal acquisition circuits; each of the signal acquisition circuits includes a first resistor, a second resistor, a capacitor and an amplifier; the first resistor and the capacitor are connected in parallel, and one end is grounded, and the other end is connected to one end of the second resistor; the other end of the second resistor is connected to a power supply; the first resistor and the second resistor are connected to the positive input end of the amplifier, and are connected to the output end of one of the friction electric self-driven sensors; the negative pole of the amplifier is connected to its output end, and the output end of the amplifier outputs a filtered and amplified neck movement signal.

[0013] Optionally, in the neck posture monitoring system based on self-driven sensing, the control unit includes a data acquisition unit, a data analysis unit, a data display unit and a health reminder unit; the data acquisition unit is used to acquire neck movement signals; the data analysis unit is used to generate neck posture according to the neck movement signals collected by the monitoring unit, and confirm whether the neck posture is an unhealthy posture; the data display unit is used to display the analysis results of the data analysis unit in a visual manner; the health reminder unit is used to issue a health reminder when the data analysis unit confirms that the neck posture is an unhealthy posture.

[0014] Optionally, in the self-driven sensing-based neck posture monitoring system, the data analysis unit includes a convolutional neural network model and / or a sliding window threshold analysis model; the convolutional neural network model is used to analyze and process the neck motion signal based on the convolutional neural network to generate the neck posture; the sliding window threshold analysis model is used to set a time window and perform threshold judgment on the signal within the time window to generate the neck posture.

[0015] Optionally, in the self-driven sensing-based neck posture monitoring system, the health reminder unit is provided with a reminder logic, and the reminder logic comprises: if the neck posture is maintained in an unfavorable posture for a period exceeding a preset maintenance time, the health manifestation unit issues a health reminder.

[0016] The neck posture monitoring system based on self-driven sensing provided by the present invention includes: a monitoring part, which is used to be attached to the human body and obtain neck movement signals based on the principle of self-driven sensing; a control part, which is used to generate neck posture according to the neck movement signals collected by the monitoring part, and confirm whether the neck posture is a bad posture. Since the monitoring part obtains neck movement signals based on the principle of self-driven sensing, the monitoring part can accurately obtain neck movement signals while having the advantages of light weight, small size, and easy wearing; at the same time, since the monitoring part is linked with the control part, the control part can conveniently and quickly obtain relevant data of the monitoring part, and control the monitoring part, which is convenient for people to adjust bad neck postures, and solves the problem that the existing neck posture monitoring device cannot provide a convenient and comfortable wearing experience, and its posture monitoring accuracy is poor. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the structure of the neck posture monitoring system based on self-driven sensing provided in this embodiment; Figure 2 A structural block diagram of the flexible data acquisition unit provided in this embodiment; Figure 3 A circuit schematic diagram of the signal acquisition circuit provided in this embodiment; Figure 4 The signal waveform diagram output by the signal acquisition circuit provided in this embodiment after inputting different sub-motion electrical signals; Figure 5 A structural block diagram of the control unit provided in this embodiment; Figure 6 A schematic diagram of the classification results of the convolutional neural network model provided in this embodiment; Figure 7 A schematic diagram of the process of sliding window threshold analysis provided in this embodiment; Figure 8 A processing flow chart of the sliding window threshold analysis model provided in this embodiment; Fig. 9 A schematic diagram of the entire process of the neck posture monitoring system using a sliding window threshold analysis model provided in this embodiment; Fig.10 A schematic diagram of the neck posture animation displayed on the monitoring tab provided in this embodiment; The descriptions of the reference numerals are as follows: 100-monitoring unit; 110-wearable braid; 120-self-driven sensor unit; 130-flexible data acquisition unit; 131-microcontroller; 132-signal collector; 133-wireless communication device; 134-power manager; 200 - control unit; 210 - data acquisition unit; 220 - data analysis unit; 230 - data display unit; 240 - health reminder unit. DETAILED DESCRIPTION

[0018] The following is a further detailed description of the self-driven sensing-based neck posture monitoring system proposed by the present invention in conjunction with the accompanying drawings and specific embodiments. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the accompanying drawings is often a part of the actual structure. In particular, the emphasis that each drawing needs to show is different, and sometimes different proportions are used.

[0019] It should be noted that the terms "first", "second", etc. in the specification, claims and drawings of the present invention are used to distinguish similar objects in order to describe the embodiments of the present invention, rather than to describe a specific order or sequence. It should be understood that the structures used in this way can be interchanged under appropriate circumstances. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] Self-driven sensing technology usually includes three mechanisms: triboelectric, piezoelectric and pyroelectric. Among them, the principle of triboelectric sensing technology is to use the magnitude of friction to produce a charge separation effect, which can generate corresponding electrical signals without the need for an external power supply, thereby realizing the measurement and monitoring of friction. As an emerging self-driven sensor, triboelectric sensors have the advantages of rich material selection, high sensitivity, small size and light weight. In particular, the single-electrode mode has the characteristics of simple structure, which is very suitable for wearable human posture monitoring.

[0021] Based on this, this embodiment provides a neck posture monitoring system based on self-driving sensing, such as Figure 1 As shown, including: The monitoring unit 100 is used to be attached to the human body and obtain neck movement signals based on the self-driven sensing principle; The control unit 200 is used to generate a neck posture according to the neck movement signal collected by the monitoring unit, and to confirm whether the neck posture is an unhealthy posture.

[0022] The neck posture monitoring system based on self-driven sensing provided in this embodiment has the advantages of light weight, small size and easy wearing while accurately acquiring neck movement signals, because the monitoring part acquires neck movement signals based on the principle of self-driven sensing. At the same time, because the monitoring part is linked with the control part, the control part can conveniently and quickly acquire relevant data of the monitoring part and control the monitoring part, which makes it convenient for people to adjust bad neck posture, and solves the problem that the existing neck posture monitoring device cannot provide a convenient and comfortable wearing experience and has poor posture monitoring accuracy.

[0023] Furthermore, in this embodiment, the monitoring part 100 includes a wearable fabric 110 and a self-driven sensing unit 120 arranged on the side of the wearable fabric 110 that contacts the human body surface; the wearable fabric 110 is used to be worn on the human neck; the self-driven sensing unit 120 includes a plurality of friction electric self-driven sensors, which are used to be attached to the human neck to obtain neck movement signals.

[0024] The neck posture monitoring system based on self-driven sensing provided in this embodiment develops a more convenient and efficient neck posture monitoring device by applying friction electric sensing technology, thereby improving the user experience and health management level.

[0025] In order to facilitate wearing and fixing, in this embodiment, the wearing braid 110 has a certain degree of stretchability, and its outer edge can be provided with an elastic silicone body, so that the wearing braid 110 can be perfectly fitted to the surface of human skin by utilizing the viscosity of the silicone body to avoid displacement of the wearing braid 110. Of course, in actual applications, the monitoring unit 100 can also be made easier to wear and fix by providing buckles or other methods, and this application does not limit this.

[0026] Wearing the braid 110 can not only provide basic support and wearing comfort, but also effectively avoid the problem of signal weakening caused by the secretion of oil, sweat and other substances on the surface of human skin, which changes the electrical properties of the interface. The specific material of the braid 110 is not limited in this application.

[0027] Furthermore, in order to make the triboelectric self-driven sensor fully contact with human skin, in this embodiment, the triboelectric self-driven sensor has a flexible substrate, and the flexible substrate is fixed to the wearable braid 110. In this way, the electrical signal (sub-motion electrical signal) obtained by the triboelectric self-driven sensor can be more accurate, so that the judgment result obtained by the control unit 200 is more accurate.

[0028] Furthermore, in order to obtain sufficient neck motion signals so that the control unit 200 can generate more accurate neck posture results, in this embodiment, a plurality of the triboelectric self-driving sensors are arranged in an array, and each of the triboelectric self-driving sensors generates its own sub-motion electrical signal. Figure 1 As shown, four friction electric self-driving sensors are arranged in an array, so that sub-motion electrical signals corresponding to four areas of the neck can be obtained. After the control unit 200 analyzes and processes the four sub-motion electrical signals, the current neck posture can be accurately obtained.

[0029] In actual application, when the triboelectric self-driving sensor contacts, moves or separates from human skin, the triboelectric self-driving sensor will generate positive or negative currents (sub-motion electrical signals) of different sizes. Afterwards, the control unit 200 analyzes and processes the sub-motion electrical signals to obtain the corresponding neck posture.

[0030] Preferably, since the self-driving sensor unit 120 is only used to obtain the neck movement signal, in order to transmit the obtained neck movement signal to the control unit 200 for analysis and processing, in this embodiment, Figure 1 As shown, the monitoring unit 100 further includes a flexible data acquisition unit 130 , and the flexible data acquisition unit 130 is used to transmit the neck movement signal acquired by the self-driving sensor unit 120 to the control unit 200 in a wireless communication manner.

[0031] In practical applications, the flexible data acquisition unit 130 can be attached to the back of the human body, or a suitable position can be selected for stably placement according to actual needs to ensure that the neck movement signal generated by the self-driving sensor unit 120 can be fully acquired and the neck movement signal can be transmitted intact to the control unit 200.

[0032] Specifically, in this embodiment, Figure 2 As shown, the flexible data acquisition unit 130 includes a microcontroller 131, a signal collector 132, a wireless communication device 133 and a power manager 134; the microcontroller 131, the signal collector 132 and the wireless communication device 133 are all connected to the power manager 134 to obtain the required voltage; the signal collector 132 and the wireless communication device 133 are all connected to the microcontroller 131 to be controlled by the microcontroller 131; the signal collector 132 can be connected to the self-driven sensing unit 120 through a flexible printed circuit board or a thin wire to obtain the neck motion signal, and filter and amplify the neck motion signal; the wireless communication device 133 is wirelessly connected to the control unit 200 to transmit the filtered and amplified neck motion signal to the control unit 200 in a wireless communication manner.

[0033] In this embodiment, the microcontroller 131, the signal collector 132, the wireless communication device 133 and the power manager 134 are all designed with low power consumption and low noise to extend the system working time and ensure its reliability. In addition, the microcontroller 131, the signal collector 132 and the wireless communication device 133 are all integrated on a flexible printed circuit board (FPCB) to minimize the volume and area of ​​the flexible data acquisition unit 130 while being able to fit the surface of human skin. Multiple interfaces can be reserved on the FPCB to facilitate the installation, replacement and upgrade of each device.

[0034] In a specific embodiment, the size of the flexible data acquisition unit 130 is only 43.86mm×31.69mm, and a 18.88mm FPCB extends from the upper left corner to connect with the self-driven sensor unit 120 to ensure the reliability and stability of signal transmission. The overall power consumption is less than 0.2 mW, which is very energy-saving.

[0035] The microcontroller 131 is the core of the entire flexible data acquisition unit 130 and is used to control processes such as signal acquisition, analog-to-digital conversion, signal processing and signal transmission.

[0036] In practical applications, the microcontroller 131 can be an ATmega328P microcontroller, which is equipped with 8 10-bit ADC pins, so as to meet the analog-to-digital conversion requirements of the sub-motion electrical signals of the four-way triboelectric self-driving sensor. In addition, the ATmega328P microcontroller also integrates SPI and serial port interfaces to ensure stable data communication with the wireless communication device 133.

[0037] The signal collector 132 is used to receive the weak sub-motion electrical signals generated by the triboelectric self-driving sensor and convert them into voltage signals that can be recognized by the microcontroller 131. It is also used to filter and amplify these signals to obtain stable and accurate voltage signals.

[0038] In this embodiment, the signal collector 132 includes a plurality of signal collection circuits, each of which is connected to a triboelectric self-driving sensor to receive the sub-motion electrical signal generated by the triboelectric self-driving sensor. For example, when there are four triboelectric self-driving sensors, the signal collector 132 is also provided with four corresponding signal collection circuits, and the signal collection circuits are connected to the triboelectric self-driving sensors one by one.

[0039] Typically, the signal acquisition circuit includes a current-to-voltage module for converting the sub-motion electrical signal (current signal) into a voltage signal recognizable by the microcontroller 131; a low-noise amplifier for amplifying the weak sub-motion electrical signal; and a signal processing module for filtering out noise and other interference signals in the sub-motion electrical signal to ensure the accuracy and stability of the data.

[0040] Specifically, in this embodiment, Figure 3 As shown, each of the signal acquisition circuits includes a first resistor R1, a second resistor R2, a capacitor C and an amplifier U; the first resistor R1 and the capacitor C are connected in parallel, and one end is grounded, and the other end is connected to one end of the second resistor R2; the other end of the second resistor R2 is connected to a power supply; the first resistor R1 and the second resistor R2 are connected to the positive input end of the amplifier U, and are connected to the output end of one of the friction electric self-driving sensors to receive the sub-motion electrical signal I output by the friction electric self-driving sensor; the negative pole of the amplifier U is connected to its output end, and the output end of the amplifier outputs the filtered and amplified neck motion signal and inputs it into the ADC pin of the microcontroller 131.

[0041] exist Figure 3 In the signal acquisition circuit shown, the first resistor R1 and the second resistor R2 can be precision resistors of 100MΩ, and the two form a voltage divider to adjust the voltage level of the amplifier U node. In this embodiment, when there is no current (sub-motion electrical signal) input, the output voltage of the signal acquisition circuit is 2.5V (baseline voltage); when there is a positive current input, the voltage output by the signal acquisition circuit will instantly rise to above 2.5V, and then fall back to the baseline, forming a positive pulse; on the contrary, when there is a negative current input, the voltage output by the signal acquisition circuit will instantly drop to below 2.5V, and then rise back to the baseline, forming a negative pulse.

[0042] Figure 4 The figure shows the changes in the potential signal output by the signal acquisition circuit of the triboelectric self-driving sensor under different actions, where a is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a resting posture, b is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a slightly lowered head posture, c is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a significantly lowered head posture, d is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a left head turn posture, e is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a right head turn posture, f is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a head-up posture, g is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a head-shaking posture, and h is the electrical signal waveform corresponding to the four triboelectric self-driving sensors in a head turning posture.

[0043] from Figure 4It can be seen that when the neck posture is different, the current (sub-motion electrical signal) output by each triboelectric self-driving sensor is different after being processed by the signal acquisition circuit.

[0044] In practical applications, the amplifier U can use the AD8608 high-precision operational amplifier, which has the characteristics of low noise and low power consumption. The four-channel CMOS structure of AD8608 effectively increases the input impedance and reduces the output impedance, thereby ensuring that the signal can be stably transmitted to the ADC pin of the microcontroller. This design greatly reduces the interference in the signal transmission process and improves the measurement accuracy and system reliability.

[0045] And, in order to further eliminate high-frequency noise, in this embodiment, a first-order low-pass filter composed of a capacitor C and a first resistor R1 is used in the signal acquisition circuit. This filtering design can effectively protect the operational amplifier and ADC from external interference, ensuring that the signal transmitted to the ADC always remains clear and stable.

[0046] The wireless communication device 133 can be specifically used to implement one or more wireless communication modes such as Bluetooth communication, NFC communication, WiFi communication, etc., so that the neck movement signal processed by the signal collector can be transmitted to the control unit 200 in a wireless communication manner.

[0047] The input end of the power manager 134 can be a +5V DC power supply, and its main function is to provide power for the microcontroller 131, the signal collector 132, the wireless communication device 133 and other electronic components.

[0048] Specifically, in actual applications, the operating voltage of the wireless communication device 133 is 3.3V, so the power manager 134 can provide a stable 3.3V DC voltage to the wireless communication device 133 through a low voltage dropout regulator; and the operating voltage of the microcontroller 131 is 5V, and a MOS tube is required for level matching between the interface of the wireless communication device 133 and the IO port of the microcontroller 134 to achieve stable communication.

[0049] Furthermore, in this embodiment, if Figure 5 As shown, the control unit 200 includes a data acquisition unit 210, a data analysis unit 220, a data display unit 230 and a health reminder unit 240; the data acquisition unit 210 is used to acquire neck movement signals; the data analysis unit 220 is used to generate neck posture according to the neck movement signals collected by the monitoring unit, and confirm whether the neck posture is a bad posture; the data display unit 230 is used to display the analysis results of the data analysis unit 220 in a visual manner; the health reminder unit 240 is used to issue a health reminder when the data analysis unit 220 confirms that the neck posture is a bad posture.

[0050] In practical applications, the control unit 200 can be a terminal with data communication and data analysis and processing functions such as a mobile phone, a tablet, a laptop computer, a smart watch, etc., and this application does not limit this. The functions of the control unit can be realized by downloading the corresponding application program on the terminal.

[0051] Specifically, in this embodiment, the data acquisition unit 210 should have a wireless communication function that matches the wireless communication device 133. For example, when the wireless communication device 133 is used for Bluetooth communication, the data acquisition unit 210 should have a Bluetooth communication function, so that it can achieve communication pairing with the wireless communication device 133, and then ensure that the neck movement signal sent by the wireless communication device 133 can be received.

[0052] Furthermore, in this embodiment, two data analysis strategies are provided, namely, the data analysis unit 220 includes a convolutional neural network model and / or a sliding window threshold analysis model; the convolutional neural network model is used to analyze and process the neck motion signal based on the convolutional neural network to generate the neck posture; the sliding window threshold analysis model is used to set the time window and perform threshold judgment on the signal within the time window to generate the neck posture.

[0053] Among them, the Convolutional Neural Network (CNN) model is a deep learning model commonly used in image and signal processing. The basic principle of CNN is to automatically extract the spatial features of the input data through the convolution layer, and use the pooling layer to reduce the dimension, so as to extract more abstract and high-level features layer by layer, and finally complete the classification task through the fully connected layer. This method is particularly suitable for complex pattern recognition and classification problems, and has strong feature learning and expression capabilities.

[0054] CNN can achieve end-to-end learning of input data through hierarchical structure and multi-layer convolution operations, so as to efficiently handle nonlinear problems. Its main advantage is that it can automatically learn data features, reduce the need for artificial feature engineering, and obtain better generalization performance on large data sets. Therefore, the convolutional neural network model used in this embodiment can accurately distinguish a variety of neck postures, such as lowering the head, raising the head, turning the head left, turning the head right, and shaking the head, after training with a large data set.

[0055] After using the convolutional neural network model to analyze and process the sub-motion electrical signals generated by multiple triboelectric self-driving sensors, the classification results are as follows: Figure 6 As shown in the figure, the classification accuracy of the convolutional neural network model can reach 99.0%, indicating that CNN has extremely high accuracy in handling such posture classification tasks. This result further verifies the reliability of CNN as a complex posture discrimination method in this example.

[0056] Although the convolutional neural network model can accurately identify seven or more neck postures, it is slightly insufficient in real time, and usually has a delay of nearly 1 second. Therefore, this embodiment also provides a sliding window threshold analysis model.

[0057] Sliding window threshold analysis is to set a time window and make a threshold judgment on the signal within the time window, so as to analyze the current posture state in real time. Figure 7 As shown in the figure, the sliding window threshold analysis starts with the collection of user data under different neck states, ensuring that the amount of data for each state reaches more than 100 groups, so as to build a representative and reliable data set. Subsequently, these data are input into the threshold analysis program, and the appropriate threshold range is divided through the algorithm, and finally a "threshold table" is generated.

[0058] In the specific threshold analysis process, the triboelectric self-driving sensor generates positive and negative pulse signals when in contact and separation, respectively. Positive pulses usually correspond to the movement of the neck, while negative pulses indicate that the neck returns to its initial state. In order to capture key posture changes, the program mainly extracts the positive pulse part of the signal. This process includes several steps: first, the baseline is determined by calculating the median of the signal; then, the standard deviation of the signal is calculated; finally, the part above the baseline plus two times the standard deviation is defined as a positive pulse signal. In order to ensure that most state-changing signals are effectively captured, the program calculates the 5% and 95% quantiles of the pulse signal, and then determines the 90% signal range and excludes the deviation value. The final generated "threshold table" records the maximum and minimum ranges of all triboelectric self-driving sensors in each state, and these data are subsequently used for real-time state analysis.

[0059] In actual application, Figure 8 As shown, the sliding window threshold analysis model first receives the data of all triboelectric self-driving sensors (in this embodiment, the four triboelectric self-driving sensors are recorded as S1, S2, S3 and S4) within a time window, and obtains the maximum and minimum values ​​of these data respectively. If these data are all within their respective baseline ranges, it indicates that the neck state has not changed significantly; if the maximum value exceeds the baseline range (baseline value ± 200), the program matches the data with the preset "threshold table" and outputs the corresponding analysis results; if the maximum value is within the baseline range and the minimum value is lower than the baseline, the program outputs "normal state", indicating that the user's neck remains in a neutral and natural position.

[0060] In this embodiment, a sliding window of 2 seconds is set to record all sensor data at the current moment and within the previous 2 seconds, and perform real-time analysis every 10ms. By comparing these data with the preset threshold, the program can quickly identify and determine the user's current neck posture.

[0061] Although the sliding window threshold analysis model has extremely high real-time performance and a delay of less than 10 ms, it can only identify basic postures such as a slight bow, a large bow, a left turn, and a right turn. For more complex neck postures, this method has limitations in its ability to distinguish. Therefore, in actual applications, users can choose the appropriate data analysis model according to actual needs.

[0062] In a specific embodiment, if Fig. 9 As shown, all the neck motion signals (sub-motion electrical signals) collected by the triboelectric self-driving sensors are processed by the flexible data acquisition unit, transmitted to the data acquisition unit via the wireless communication device (Bluetooth), and then sent to the sliding window threshold analysis model for analysis and processing. After that, the analyzed neck posture is presented to the user in a visual manner.

[0063] Furthermore, in this embodiment, the data display unit 230 generally has a visual interface, so that the analysis results can be presented to the user in a graphical form, including sensor waveforms, posture monitoring results, and animations of posture changes, so that the user can intuitively understand the changes in their neck posture. For example, when the control unit 200 is a mobile phone, the data display unit 230 can be a mobile phone display screen, so that the current neck posture can be intuitively and clearly obtained in a visual manner.

[0064] Preferably, in order to facilitate users to obtain more information, in this embodiment, the data display unit 230 is provided with a plurality of main tab pages, such as a log tab page, a data tab page and a monitoring tab page.

[0065] The log tab is used to display the logs generated by the neck posture monitoring system. In this tab interface, users can view various events that occurred in the application at different time points, including but not limited to changes in wireless communication connection status, mode switching, and other system status updates. Through the log interface, users can track the application's operation history in detail, helping users understand the device's operation process and perform troubleshooting.

[0066] The data tab allows users to select data analysis methods and display the relevant data processing process. This tab interface presents the historical data collected by the device to the user and provides two data analysis methods for the user to choose: convolutional neural network and threshold analysis based on sliding window. Users can choose the appropriate analysis method according to actual needs. The selected method will be used to process the collected data to help users obtain and evaluate the status information of the neck posture in real time.

[0067] The monitoring tab is used to display the current monitoring results of the neck posture in real time, including but not limited to the three-dimensional model of the neck posture, the judgment result of whether the neck posture is a bad posture, the working status of each triboelectric self-driving sensor, etc. This tab interface is divided into three main parts: The upper part can display real-time status animation, which dynamically reflects the user's current neck posture and provides visual feedback in the form of animation. According to different modes, different modes and different postures correspond to animation diagrams, such as Fig.10 , where a to h represent normal posture, slight bowing, large bowing, raising head, turning left, turning right, turning head and shaking head in working state respectively, and i to l represent normal posture, bowing head, turning left and turning right in driving state respectively. If the user maintains an unhealthy neck posture for more than the preset time threshold, a reminder pop-up window will pop up on the upper part of the interface, accompanied by sound and vibration, to remind the user to adjust the posture and maintain a healthy neck state. Mode switching can also be achieved by clicking on the upper display area. These animated diagrams provide users with intuitive visual feedback to help them identify and correct poor neck posture, ensure that the correct posture is maintained in different scenarios, and help prevent neck fatigue and related health problems.

[0068] The middle area can display the mode information selected by the user (working mode or driving mode), as well as the real-time information of the neck posture and posture maintenance time related to the current mode. This part of the information display allows users to understand their neck health status at any time and make timely adjustments.

[0069] The lower part consists of multiple signal waveform displays, the number of which is consistent with the number of triboelectric self-driving sensors, which intuitively display the real-time data of the triboelectric self-driving sensors in the form of waveforms. This function not only helps users understand the changes in neck posture, but also provides professionals with detailed posture monitoring data for further analysis and diagnosis.

[0070] In addition, the bottom of the monitoring tab can also be equipped with multiple function buttons, such as pause, clear screen and export. The "Pause" button can be used to temporarily stop the display of data to help users focus on a specific moment; the "Clear Screen" button is used to clear the currently displayed data so that users can restart monitoring. The "Export" button allows users to export data, support further analysis of the data, and can share the data to social media platforms so that users can save and share monitoring results.

[0071] Furthermore, in this embodiment, the health reminder unit is provided with a reminder logic, and the reminder logic includes: if the neck posture is maintained in an unfavorable posture for a time exceeding a preset maintenance time, the health manifestation unit issues a health reminder.

[0072] Among them, the preset maintenance time can be set according to user needs, so that the neck posture can be better adjusted.

[0073] In actual applications, the timer function inside the terminal (such as a mobile phone) can be used. When the user is detected to maintain a certain bad posture for a long time (such as lowering the head, raising the head, turning left or right, etc. for more than the set time threshold), the program will automatically make a sound and vibrate to remind the user to adjust the posture in time to prevent neck fatigue or injury caused by improper posture. This function is designed to help users develop good posture habits through timely feedback, thereby reducing the risk of neck diseases.

[0074] Preferably, considering that the neck posture may be maintained differently in different scenarios, such as in working or driving state, the user may need to maintain a straight-looking state for a long time. Therefore, in this embodiment, two monitoring modes with different parameters are set: one is the working mode, which focuses on monitoring and correcting the user's sitting posture in an office environment; the other is the driving mode, which is specially designed for posture monitoring while driving, helping the driver to maintain a good driving posture and reduce fatigue or safety hazards caused by bad posture. Both modes can automatically adjust the monitoring parameters and reminder methods according to the specific scene requirements, improving the user experience while better meeting the posture monitoring needs in different environments.

[0075] The neck posture monitoring system based on self-driven sensing provided in this embodiment collects neck movement signals through flexible devices of triboelectric effect, and transmits the signals to the control unit through wireless communication. The control unit performs real-time analysis on the collected neck movement signals through built-in algorithms or algorithm models, thereby realizing real-time and accurate monitoring and feedback of neck posture.

[0076] Among them, the triboelectric self-driven sensor can be reliably positioned while fitting the neck curve through a flexible substrate, thereby ensuring the accuracy and stability of the collected electrical signals; at the same time, by wearing the woven fabric and the triboelectric self-driven sensor's light weight and small size, it can provide a stable and comfortable wearing experience, and can be attached to other parts of the body as needed.

[0077] In addition, the convolutional neural network model accurately classifies and predicts the neck posture, effectively coping with various posture changes; the sliding window threshold analysis model monitors the degree of bending of the neck angle and the duration of maintenance. The two analysis methods can not only analyze the current user's neck posture with high accuracy, but also meet the user's neck posture monitoring needs in various scenarios.

[0078] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other. In addition, the different parts between the various embodiments can also be used in combination with each other, and the present invention is not limited to this.

[0079] The neck posture monitoring system based on self-driven sensing provided in this embodiment includes: a monitoring unit, which is used to be attached to the human body and obtain neck movement signals based on the principle of self-driven sensing; a control unit, which is used to generate neck posture according to the neck movement signals collected by the monitoring unit, and confirm whether the neck posture is a bad posture. Since the monitoring unit obtains neck movement signals based on the principle of self-driven sensing, the monitoring unit can accurately obtain neck movement signals while having the advantages of light weight, small size, and easy to wear; at the same time, since the monitoring unit is linked with the control unit, the control unit can conveniently and quickly obtain relevant data of the monitoring unit and control the monitoring unit, which is convenient for people to adjust bad neck postures, and solves the problem that the existing neck posture monitoring device cannot provide a convenient and comfortable wearing experience, and its posture monitoring accuracy is poor.

[0080] The above description is only a description of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any changes or modifications made by a person skilled in the art in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. A neck posture monitoring system based on self-driven sensing, characterized in that: include: The monitoring part is used to be attached to the human body and obtain the neck movement signal based on the self-driven sensing principle; The control unit is used to generate the neck posture according to the neck movement signal collected by the monitoring unit, and confirm whether the neck posture is a bad posture.

2. The neck posture monitoring system based on self-driven sensing according to claim 1 is characterized in that: The monitoring part includes a wearable fabric and a self-driven sensing unit arranged on the side of the wearable fabric in contact with the human body surface; the wearable fabric is used to be worn on the human neck; the self-driven sensing unit includes a plurality of friction electric self-driven sensors, which are used to be attached to the human neck to obtain neck movement signals.

3. The neck posture monitoring system based on self-driven sensing according to claim 2 is characterized in that: The triboelectric self-driven sensor has a flexible substrate, and the flexible substrate is fixed to the wearable braid.

4. The neck posture monitoring system based on self-driven sensing according to claim 2 is characterized in that: The plurality of triboelectric self-driving sensors are arranged in an array, and each of the triboelectric self-driving sensors generates its own sub-motion electrical signal.

5. The neck posture monitoring system based on self-driven sensing according to claim 2 is characterized in that: The monitoring unit further includes a flexible data acquisition unit, and the flexible data acquisition unit is used to transmit the neck movement signal acquired by the self-driving sensor unit to the control unit in a wireless communication manner.

6. The neck posture monitoring system based on self-driven sensing according to claim 5 is characterized in that: The flexible data acquisition unit includes a microcontroller, a signal collector, a wireless communication device and a power manager; the microcontroller, the signal collector and the wireless communication device are all connected to the power manager to obtain the required voltage; the signal collector and the wireless communication device are both connected to the microcontroller to be controlled by the microcontroller; the signal collector is connected to the self-driven sensing unit to obtain the neck motion signal and filter and amplify the neck motion signal; the wireless communication device is wirelessly connected to the control unit to transmit the filtered and amplified neck motion signal to the control unit by wireless communication.

7. The neck posture monitoring system based on self-driven sensing according to claim 6, characterized in that: The signal collector includes multiple signal acquisition circuits; each of the signal acquisition circuits includes a first resistor, a second resistor, a capacitor and an amplifier; the first resistor and the capacitor are connected in parallel, and one end is grounded, and the other end is connected to one end of the second resistor; the other end of the second resistor is connected to a power supply; the first resistor and the second resistor are connected to the positive input end of the amplifier, and are connected to the output end of one of the friction electric self-driven sensors; the negative pole of the amplifier is connected to its output end, and the output end of the amplifier outputs a filtered and amplified neck movement signal.

8. The neck posture monitoring system based on self-driven sensing according to claim 1, characterized in that: The control unit includes a data acquisition unit, a data analysis unit, a data display unit and a health reminder unit; the data acquisition unit is used to acquire neck movement signals; the data analysis unit is used to generate neck posture according to the neck movement signals collected by the monitoring unit, and confirm whether the neck posture is an unhealthy posture; the data display unit is used to display the analysis results of the data analysis unit in a visual manner; the health reminder unit is used to issue a health reminder when the data analysis unit confirms that the neck posture is an unhealthy posture.

9. The neck posture monitoring system based on self-driven sensing according to claim 8, characterized in that: The data analysis unit includes a convolutional neural network model and / or a sliding window threshold analysis model; the convolutional neural network model is used to analyze and process the neck motion signal based on the convolutional neural network to generate the neck posture; The sliding window threshold analysis model is used to set a time window and perform threshold judgment on the signal within the time window to generate the neck posture.

10. The neck posture monitoring system based on self-driven sensing according to claim 8, characterized in that: The health reminder unit is provided with a reminder logic, and the reminder logic includes: if the neck posture is maintained for a time exceeding a preset maintenance time as a bad posture, the health manifestation unit issues a health reminder.

Images