A muscle recognition method based on posture division

Abstract

The present invention belongs to the field of myoelectric signal segmentation, and specifically relates to a muscle state identification method based on posture segmentation, comprising: synchronously collecting motion data and myoelectric signals during exercise; segmenting the motion data according to the posture movements during exercise to obtain multiple time nodes; segmenting the myoelectric signals using the multiple time nodes and identifying and analyzing muscle states. The purpose of the present invention is to utilize the temporal changes of posture movements corresponding to different stages of motion data to obtain time nodes for segmenting myoelectric signals, thereby temporally dividing the tested muscle group into the agonist and antagonist muscle states, extracting the myoelectric signals in the agonist state, and identifying and analyzing the corresponding muscle states.

Description

Technical Field

[0001] The present invention belongs to the field of myoelectric signal segmentation, and in particular relates to a muscle state recognition method based on posture segmentation. Background Art

[0002] Fatigue status assessment is of great significance for sports training and rehabilitation treatment; determining fatigue status plays an important role in muscle fatigue research, spasticity assessment research, etc.

[0003] Fatigue has complex mechanisms. Existing fatigue assessment scales, while convenient, are susceptible to subjective influences, making accurate evaluations inaccurate. While biochemical markers and near-infrared spectroscopy can detect fatigue-related indicators, these methods are invasive, require specialized procedures, and lack real-time performance. Consequently, they are limited to use in hospitals or specialized laboratories, making them impractical.

[0004] SEMG (surface electromyography) has been widely used in fatigue research, particularly in muscle fatigue, in recent years due to its non-invasive and real-time nature. In studies of static muscle contraction fatigue, the consensus is that SEMG's time-domain characteristics increase and its frequency-domain characteristics decrease with increasing fatigue. However, in studies of dynamic muscle contraction fatigue, the increasing nonlinearity and nonstationarity of EMG signals introduce uncertainties, and there is currently no consensus on this topic.

[0005] During static contractions, SEMG is typically segmented using a moving window method to calculate fatigue indicators. During dynamic contractions, SEMG is typically segmented using a feature threshold method to identify active segments. However, existing research has not analyzed the effectiveness of feature threshold methods for assessing muscle fatigue during rehabilitation exercises, nor has it analyzed methods for studying muscle fatigue during rehabilitation exercises. Furthermore, the commonly used feature threshold method cannot segment the EMG signal corresponding to when a muscle is acting as an active muscle. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a muscle state recognition method capable of extracting the required target myoelectric signal therefrom.

[0007] To achieve the above-mentioned purpose and other related purposes, the present invention provides a muscle state recognition method based on posture division, including: synchronously collecting motion data and electromyographic signals during the movement process; dividing the motion data according to the posture movements during the movement process to obtain multiple time nodes; dividing the electromyographic signals through the multiple time nodes and identifying and analyzing the muscle state.

[0008] According to one embodiment of the present invention, the step of dividing the motion data according to the posture movements during the motion process to obtain multiple time nodes includes: dividing the motion data according to the time corresponding to the preset motion process to obtain the motion data of the actual motion process; dividing the motion data of the actual motion process twice, when the motion data of the actual motion process corresponding to L consecutive time points are within a first threshold range, the first time point of the L time points is used as the starting point; when the motion data of the actual motion process corresponding to L consecutive time points are within a second threshold range, the first time point of the L time points is used as the end point.

[0009] According to an embodiment of the present invention, the motion data is the angular velocity or acceleration of the limb movement during the motion process.

[0010] According to an embodiment of the present invention, the step of synchronously collecting motion data and electromyographic signals during the motion process includes: performing denoising processing on the electromyographic signal data by using a wavelet threshold denoising algorithm.

[0011] According to an embodiment of the present invention, the step of synchronously collecting motion data and electromyographic signals during the motion process further includes: filtering the motion data using a Butterworth low-pass filtering algorithm.

[0012] According to an embodiment of the present invention, the electromyographic signal includes an agonist muscle electromyographic signal and an antagonist muscle electromyographic signal.

[0013] According to one embodiment of the present invention, the electromyographic signal corresponding to the starting point and the end point is the active muscle electromyographic signal. By dividing the electromyographic signal into multiple cycles, the active muscle electromyographic signal area is extracted, and the change trend of the active muscle electromyographic signal in each area is observed and analyzed, so as to judge the fatigue state of the muscle.

[0014] A muscle state recognition system based on posture division, comprising:

[0015] Information acquisition module, used to synchronously collect motion data and electromyographic signals during exercise;

[0016] An information extraction module, configured to divide the motion data according to the posture and action during the motion process to obtain a plurality of time nodes;

[0017] The information division module is used to divide the electromyographic signal according to the multiple time nodes and identify and analyze the corresponding muscle states.

[0018] A muscle state recognition device based on posture division includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of any one of the above methods are implemented.

[0019] A computer-readable medium stores instructions thereon, wherein the instructions are loaded by a processor and executed by any one of the methods described above.

[0020] The technical effect of the present invention is that by utilizing the time changes of different stages of motion data corresponding to posture movements, time nodes are obtained to divide the electromyographic signals, and the agonist muscle state and antagonist muscle state of the measured muscle group are divided accordingly in time, the electromyographic signals of the agonist muscle state are extracted, and analyzed, and then the state of the muscle is identified, which solves the problem in previous studies that the electromyographic signals of the agonist muscle could not be divided, and the state of the muscle is better identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a muscle state recognition method based on posture division provided by the present invention;

[0022] Figure 2 This is a schematic diagram of the arm motion state of a specific embodiment provided by the present invention;

[0023] Figure 3 This is a graph showing the time variation of angular velocity according to a specific embodiment of the present invention;

[0024] Figure 4 This is a time comparison curve diagram of angular velocity and electromyographic signal of a specific embodiment provided by the present invention;

[0025] Figure 5 This is an IEMG curve diagram of a specific embodiment provided by the present invention;

[0026] Figure 6 A flow chart of a muscle state recognition system based on posture classification provided by the present invention;

[0027] Figure 7 This is a structural schematic diagram of a muscle state recognition device based on posture division provided by the present invention. DETAILED DESCRIPTION

[0028] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.

[0029] See also Figure 1-7It should be noted that the diagrams provided in this embodiment are merely schematic illustrations of the basic concept of the present invention. Therefore, the diagrams only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0030] The embodiment of the present application can divide the electromyographic signals based on the time changes of the motion data corresponding to the posture movements corresponding to the motion process, obtain multiple time nodes by presetting the motion data corresponding to the posture movements during the motion process, divide the electromyographic signals according to the multiple time nodes, extract the required target electromyographic signals and identify and analyze the corresponding muscle states.

[0031] In the study of muscle dynamic contraction fatigue, researchers usually use the characteristic threshold method to divide SEMG into active segments, and then calculate the relevant indicators of each active segment in an exercise cycle, and characterize the occurrence process of muscle fatigue by analyzing the changes of each indicator as the number of exercises increases. However, the active segment SEMG divided by the characteristic threshold method includes the muscle as the agonist muscle and the antagonist muscle. When the muscle is the agonist muscle, it is the main way to cause muscle fatigue, and the characteristic threshold method is difficult to divide the SEMG corresponding to the agonist muscle. Therefore, the present invention proposes to divide SEMG by movement posture, and separate the sEMG part of the muscle as the agonist muscle, so as to realize the analysis of the patient's muscle fatigue during rehabilitation exercise, and realize real-time monitoring of the fatigue state of the patient's muscle in subsequent research, to provide assistance for the patient's accurate rehabilitation.

[0032] The muscle state recognition method based on posture division is applied to one or more electronic devices, which are devices that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Their hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0033] The electronic device may be any electronic product that can interact with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.

[0034] The electronic device may further include a network device and / or a user device, wherein the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of hosts or network servers.

[0035] The network where the electronic device is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0036] Example 1

[0037] See Figure 1 The schematic diagram of the muscle state recognition method process shown in FIG. 1 shows a muscle state recognition method based on posture division, which includes:

[0038] Step S10, the step of synchronously collecting motion data and electromyographic signals during the motion process includes:

[0039] Step S11: To ensure accurate EMG signal acquisition, wipe the surface of the right biceps brachii with 95% alcohol before the experiment to remove sweat and stains from the skin and reduce impedance. Wear the developed muscle fatigue data acquisition and transmission system device, with the data transmission module worn on the right wrist, the EMG signal acquisition module placed on the belly of the biceps brachii, and the angular velocity data acquisition module placed on the outside of the upper and lower arms, each secured with Velcro.

[0040] Step S12: The experimenter sits upright with his or her upper body straight and his or her right arm pointing downward with the palm facing forward. Instruct the experimenter on precautions, turn on the system, connect the computer to the local area network of the data acquisition and transmission system, and start the network debugging assistant to prepare for data reception.

[0041] Step S13, as Figure 2 As shown, the experimenter holds a 5kg dumbbell, keeps the upper arm still, bends the arm so that the forearm is in a horizontal position, then extends the arm so that the forearm is vertically downward, and after staying still for 3 seconds, the arm bends again. That is, the experimenter continues the cycle of arm bending - arm extension - stillness until the forearm can no longer bend to a horizontal position, at which point the experiment ends.

[0042] Step S14: collecting and saving angular velocity data and corresponding electromyographic signals.

[0043] The collected EMG signals were also denoised using a wavelet threshold denoising algorithm. After testing and analysis, the db4 wavelet basis function was selected, with a decomposition layer of four. A fixed threshold method was used to calculate the threshold, and a soft threshold function was used for noise processing. The wavelet threshold denoising algorithm suppresses signal noise while preserving signal breakpoints. The resulting filtered EMG signal has a relatively stable baseline.

[0044] A Butterworth filter is a type of electronic filter, also known as a maximally flat filter. The characteristic of a Butterworth filter is that its frequency response curve is maximally flat within the passband, with no ripples, while it gradually decreases to zero in the stopband.

[0045] A first-order Butterworth filter has a rolloff rate of 6 dB per octave and 20 dB per decade. A second-order Butterworth filter has a rolloff rate of 12 dB per octave, a third-order Butterworth filter has a rolloff rate of 18 dB per octave, and so on. The Butterworth filter has a monotonically decreasing amplitude versus angular frequency, and is the only filter whose amplitude versus angular frequency curve maintains the same shape regardless of filter order. However, higher-order filters exhibit a faster amplitude rolloff in the stopband. Other filters with higher-order amplitude versus angular frequency have different shapes than those with lower order amplitude versus angular frequency.

[0046] During angular velocity data collection, the collected angular velocity data may contain glitches due to factors such as variations in the position of the posture sensor and varying degrees of muscle contraction. This can hinder feature extraction and motion posture classification, necessitating filtering of the angular velocity data. A Butterworth function is used, with an 8th-order, low-pass filter and a cutoff frequency of 20Hz, to smooth the angular velocity data. The filtered data curve is smoother, with glitches effectively reduced, facilitating feature extraction and motion posture classification.

[0047] Step S20, dividing the motion data according to the postures and actions during the motion process to obtain a plurality of time nodes, includes:

[0048] There are different posture movements during the movement process, and each posture movement has a corresponding time change node. The required motion data segment and its corresponding electromyographic signal are obtained through the time node corresponding to the posture movement.

[0049] Step S21, dividing the motion data according to the time corresponding to the preset motion process to obtain motion data of the actual motion process;

[0050] The motion data is divided into three types of motion according to the postures and actions in the motion of the arm:

[0051] Arm static state: When the arm is in the static state, the forearm is kept still. However, because the forearm cannot be completely still, the angular velocity data in this stage should fluctuate around zero.

[0052] Arm bending: When the arm is bent, the forearm moves centripetally with the elbow joint as the center. Depending on the installation method of the attitude sensor, the angular velocity data at this stage should be positive.

[0053] Arm extension: When the arm is extended, the forearm performs eccentric motion with the elbow joint as the center. According to the installation method of the attitude sensor, the angular velocity data in this stage should be negative.

[0054] like Figure 3 As shown, the motion process is a repetitive action, and the corresponding data characteristics show periodic changes. The CC segment is the motion data corresponding to the motion process within a cycle.

[0055] As can be seen, the data for the CA segment remains near zero, indicating that the arm is stationary during this phase. The data for the AC segment begins to fluctuate, changing from positive to negative values, indicating that this phase corresponds to actual movement, including arm flexion and extension. The data for the AB segment is positive, and point A is the turning point where the angular velocity changes from zero to positive, consistent with the characteristic change from stationary to flexed arm motion. This indicates that this phase is arm flexion. The data for the BC segment is negative, and point B is the turning point where the angular velocity changes from positive to negative, consistent with the characteristic change from flexion to extension. This indicates that this phase is arm extension.

[0056] Since the arm cannot be in an ideal state when it is still, the SEMG signal will have certain fluctuations. In order to obtain the electromyographic signal corresponding to the motion data of the actual motion process, the motion data corresponding to the AC segment is analyzed and divided twice.

[0057] Step S22 sets a judgment length L and divides the motion data of the actual motion process into two parts. When L consecutive angular velocity data are all greater than zero, the time corresponding to the first angular velocity data is used as the starting point. When L consecutive angular velocity data are all less than zero, the time corresponding to the first angular velocity data is used as the ending point. Thus, time nodes A and B are obtained.

[0058] It should be noted that, in this embodiment, the turning point between different stages of the motion process is exactly the corresponding first time point obtained, and the method proposed in the present invention is aimed at a variety of different situations and various aspects of the field.

[0059] Step S30: dividing the electromyographic signal according to the multiple time nodes and identifying and analyzing the corresponding muscle states.

[0060] like Figure 4The figure shows the electromyographic signal and angular velocity analysis diagram, marking the start and end time points M and N of the active segment of the electromyographic signal, as well as the angular velocity corresponding to different stages of the movement process CA, AC and the acquisition time nodes A and B.

[0061] Identify and analyze the muscle states corresponding to different areas of the divided electromyographic signals.

[0062] CM: In this stage, the arm is at rest, the load force is mainly provided by the bones, the biceps brachii is relaxed, and the SEMG is at rest.

[0063] MA: In this stage, the brain's nervous system begins to control the contraction of the biceps to complete the movement. However, because the electromyographic signal is transmitted quickly, the SEMG has begun to become active even though the arm has not yet begun to bend. It should also be noted that the electromyographic signal in the MA stage is in the rising stage and the biceps begins to contract.

[0064] Since the CA segment is the state of the arm at rest, the CM and MA are not the areas where the electromyographic signals need to be obtained.

[0065] AB: During this stage, the biceps brachii continuously and actively contracts to drive the forearm to do concentric motion around the elbow joint. During this stage, the biceps brachii actively contracts, the SEMG is in an active state, and its change trend is consistent with that of the angular velocity, both increasing first and then decreasing before the arm is extended.

[0066] From the above, we can see that the AB segment is the electromyographic signal area of ​​the agonist muscle.

[0067] BN: In this stage, the arm begins to extend, the triceps begins to contract, and the forearm performs eccentric movement around the elbow joint. The biceps is in passive relaxation, and the SEMG is still active, but its SEMG activity is lower than that of the AB segment.

[0068] From the above, we can see that the BN segment is the antagonistic muscle EMG signal area.

[0069] NC: Although the arm is still extending at this stage, it is due to the gravity of the load, and the biceps has been relaxed to the point where the contraction is no longer obvious, so the SEMG is in a resting state.

[0070] Therefore, the agonist muscle EMG signal is the EMG signal corresponding to the AB segment, and the antagonist muscle EMG signal is the EMG signal corresponding to the BN segment. The EMG signals of different segments can be selected for the next step of analysis and processing according to the needs. In this embodiment, the EMG signal area of ​​the agonist muscle is obtained for subsequent muscle fatigue research.

[0071] IEMG (Integrated Electromyogram) refers to the sum of the areas under the curve of all myoelectric signals after rectification and filtering per unit time. It reflects the total discharge of the MUs (Mount Units) involved in the activity within a certain period of time. That is, under the premise of constant time, the size of its value reflects, to a certain extent, the number of MUs involved in the work and the discharge size of each MU. The IEMG calculation formula is as follows:

[0072]

[0073] Among them, X i (i=1, 2, ..., N) is the time series of the electromyographic signal, and N is the selected frame length.

[0074] In a specific embodiment, as the degree of muscle fatigue increases, the IEMG fluctuation amplitude will also increase, so the degree of muscle fatigue can be judged based on the IEMG value. Figure 5 As shown, as the number of corresponding exercises increases, the IEMG value also increases, so it can be concluded that the degree of muscle fatigue also increases. The calculated slope is 0.000481, and the IEMG sensitivity is 67.61%.

[0075] It should be noted that this embodiment only divides the electromyographic signals within one cycle. The action recognition method based on posture division provided by the present invention can be applied to electromyographic signals within any cycle. By dividing the electromyographic signals of multiple cycles, the active muscle electromyographic signal areas are extracted, and the changing trends of the active muscle electromyographic signals in each area are observed and analyzed, thereby judging the fatigue state of the muscles.

[0076] It should be noted that, in the present invention, in order to ensure the security of data, the data involved can be deployed on the blockchain to prevent the data from being maliciously tampered with.

[0077] It should be noted that the step division of the various methods above is only for the purpose of clear description. During implementation, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they contain the same logical relationship, they are all within the scope of protection of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this patent.

[0078] Example 2

[0079] like Figure 6 As shown, this embodiment discloses a muscle state recognition system based on posture division, including:

[0080] Information acquisition module, used to synchronously collect motion data and electromyographic signals during exercise;

[0081] An information extraction module, configured to divide the motion data according to the posture and action during the motion process to obtain a plurality of time nodes;

[0082] The information division module is used to divide the electromyographic signal according to the multiple time nodes and identify and analyze the corresponding muscle states.

[0083] It should be noted that, when actually implemented, the aforementioned functional modules can be fully or partially integrated into a single physical entity, or physically separated. Furthermore, these modules can all be implemented in the form of software called by a processing element; or all be implemented in the form of hardware; or some modules can be implemented in the form of software called by a processing element, while others are implemented in the form of hardware. Furthermore, these modules can all or partly be integrated together, or implemented independently. The processing element described herein can be an integrated circuit with signal processing capabilities. During implementation, some or all of the steps of the aforementioned method, or the aforementioned functional modules, can be completed by hardware integrated logic circuits in the processor element or by software instructions.

[0084] Example 3

[0085] like Figure 7 As shown, this embodiment discloses a muscle state recognition device based on posture division, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0086] The electronic device may include a memory, a processor, and a bus, and may further include a computer program stored in the memory and executable on the processor.

[0087] Among them, the memory includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (for example, SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory can be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory can also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Furthermore, the memory can also include both an internal storage unit of the electronic device and an external storage device. The memory can be used not only to store application software and various types of data installed in the electronic device, but also to temporarily store data that has been output or is to be output.

[0088] In some embodiments, the processor may be composed of an integrated circuit, such as a single packaged integrated circuit or a combination of multiple packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor is the control core (Control Unit) of the electronic device, connecting the various components of the entire electronic device using various interfaces and circuits. It executes or runs programs or modules stored in the memory and calls data stored in the memory to perform various functions of the electronic device and process data.

[0089] The processor executes the operating system of the electronic device and various installed application programs. The processor executes the application programs to implement the steps in the above embodiment.

[0090] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0091] The above-mentioned integrated unit implemented in the form of a software functional module can be stored in a computer-readable storage medium. The above-mentioned software functional module is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, computer equipment, or network equipment, etc.) or a processor to perform part of the functions of various embodiments of the present invention.

[0092] To sum up, the technical effect of the present invention is that by utilizing the time changes of different stages of motion data corresponding to posture movements, time nodes are obtained to divide the electromyographic signals, and the agonist muscle state and antagonist muscle state of the measured muscle group are divided accordingly in time, the electromyographic signals of the agonist muscle state are extracted, and analyzed, and then the state of the muscle is identified, which solves the problem in previous studies that the electromyographic signals of the agonist muscle cannot be divided, and the state of the muscle is better identified.

[0093] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0094] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A muscle state recognition method based on posture division, characterized in that: include: Synchronously collect motion data and electromyographic signals during exercise; The motion data is divided according to the posture and action during the motion process to obtain multiple time nodes, and the steps include: dividing the motion data according to the time corresponding to the preset motion process to obtain motion data of the actual motion process; performing secondary division on the motion data of the actual motion process, when the motion data of the actual motion process corresponding to L consecutive time points are within a first threshold range, the first time point of the L time points is used as the starting point; when the motion data of the actual motion process corresponding to L consecutive time points are within a second threshold range, the first time point of the L time points is used as the ending point; The electromyographic signal is divided by the multiple time nodes to divide the electromyographic signal into an agonist muscle electromyographic signal and an antagonist muscle electromyographic signal, and the muscle state is analyzed based on the agonist muscle electromyographic signal; wherein the electromyographic signal corresponding to the starting point and the end point is the agonist muscle electromyographic signal.

2. The muscle state recognition method according to claim 1, characterized in that: The motion data is the angular velocity or acceleration of the limb movement during the motion process.

3. The muscle state recognition method according to claim 1, characterized in that: The step of synchronously collecting motion data and electromyographic signals during the motion process includes: performing denoising processing on the electromyographic signal data by using a wavelet threshold denoising algorithm.

4. The muscle state recognition method according to claim 1, characterized in that: The step of synchronously collecting motion data and electromyographic signals during the motion process further includes: using a Butterworth low-pass filter to filter the motion data.

5. A muscle state recognition system based on posture division, characterized in that: include: Information acquisition module, used to synchronously collect motion data and electromyographic signals during exercise; An information extraction module is configured to divide the motion data according to the posture and action during the motion process to obtain multiple time nodes, including: dividing the motion data according to the time corresponding to the preset motion process to obtain motion data of the actual motion process; performing a secondary division on the motion data of the actual motion process, and when the motion data of the actual motion process corresponding to L consecutive time points are within a first threshold range, the first time point of the L time points is used as the starting point; when the motion data of the actual motion process corresponding to L consecutive time points are within a second threshold range, the first time point of the L time points is used as the ending point; An information division module is used to divide the electromyographic signals according to the multiple time nodes and identify and analyze the corresponding muscle states, so as to divide the electromyographic signals into agonist muscle electromyographic signals and antagonist muscle electromyographic signals, and analyze the muscle states based on the agonist muscle electromyographic signals; wherein the electromyographic signals corresponding to the starting point and the end point are the agonist muscle electromyographic signals.

6. A muscle state recognition device based on posture division, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable medium, characterized in that Instructions are stored thereon, and the instructions are loaded by a processor to execute the method according to any one of claims 1 to 4.

Images