A multi-modal strength training assistance method and system

Abstract

The present invention discloses a multi-modal strength training assistance method and system. The method includes: obtaining the myoelectric signal and posture signal of the user during the strength training process, identifying the starting point of the action, and determining the strength training information according to the extracted myoelectric signal characteristics and posture signal characteristics, where the strength training information includes the action type and the muscle group force intensity; obtaining the brain wave signal of the user during the strength training process, extracting the frequency domain signal characteristics according to the identified starting point of the action, and further determining the brain attention intensity information according to the frequency domain signal characteristics; based on the strength training information and the brain attention emphasis information, analyzing in combination with the actual training strength, and feeding back the strength training result to the user. The present invention detects the mental state of the user and helps the user enter the focused mode through visual feedback, can automatically identify and correct common fitness actions, and improves the correctness and efficiency of the user's strength training actions.

Description

Technical Field

[0001] The present invention relates to the technical field of strength training, and more particularly, to a multi-modal strength training assistance method and system. Background Art

[0002] With the development and progress of society, people have a higher pursuit of a strong body and posture. Strength training can help with weight loss, improve balance, and prevent falls, which is crucial for both physical and mental health. In recent years, free weight training has become an attractive exercise method that can increase muscle strength, muscle mass, and joint strength. Free weight training is widely used in fitness and exercise because it does not require too much equipment. For example, a single dumbbell can be used for various exercises.

[0003] In fact, although everyone has different exercise purposes, the common goal of "efficient training" is the same. In strength training, there are many strength training conditions according to the involved muscles, the equipment used, the rhythm of the movements, the duration, and the complexity. Moreover, correct techniques and postures are crucial for avoiding potential injuries caused by strength training. On the other hand, without a correct posture with mental concentration, the training effect cannot be guaranteed. In strength training, a mentally concentrated exercise can enhance the activation of muscle fibers in all directions, known as the "brain-muscle connection", which is more important than the weight a person can lift, more valuable than the proficient use of equipment, and even more critical than the repetitive rhythm.

[0004] Research shows that when the muscle contracts, concentrating on the target muscle can significantly promote the generation of force. Therefore, mental concentration is undoubtedly crucial for effective training. When the muscle of an organism contracts, a weak current is generated, and this bioelectrical signal can be captured through myoelectric electrodes attached to the surface of the organism. Moreover, this signal is related to the intensity of the body movement, and electromyography can be used to record the muscle load of the organism.

[0005] Brain waves are electrical signals generated by the cerebral cortex during brain activities. Electroencephalograms can record the changes in the electrical signals of brain activities, thereby inferring the state of the brain. The monitoring of electroencephalograms is widely used in clinical practice. Current research shows that electroencephalograms can be divided into four important frequency bands: δ (1 - 3 Hz), θ (4 - 7 Hz), α (8 - 13 Hz), and β (14 - 30 Hz). Different bands represent different brain activity characteristics. For example, when the human body is relaxed, the proportion of α waves increases, while when the human body is concentrating, the proportion of β waves increases.

[0006] In the prior art, although emerging online platforms and mobile applications can guide and track strength training, there is currently no practical method to utilize the brain-muscle connection for monitoring and assisting the exercise situation of users. Summary of the Invention

[0007] The object of the present invention is to overcome the defects of the above-mentioned prior art, and provide a multi-modal strength training assistance method and system, which is a new technical solution for assisting strength training using wearable electroencephalogram sensors and electromyogram sensors.

[0008] According to the first aspect of the present invention, there is provided a multi-modal strength training assistance method. The method includes the following steps:

[0009] Obtain the electromyogram signal and posture signal of the user during the strength training process, identify the starting point of the action, and determine the strength training information according to the extracted electromyogram signal characteristics and posture signal characteristics, where the strength training information includes the action type and the muscle group exertion intensity;

[0010] Obtain the electroencephalogram signal of the user during the strength training process, extract the frequency domain signal characteristics according to the identified starting point of the action, and further determine the brain attention intensity information according to the frequency domain signal characteristics;

[0011] Based on the strength training information and the brain attention emphasis information, analyze in combination with the actual training strength, and feedback the strength training result to the user.

[0012] According to the second aspect of the present invention, there is provided a multi-modal strength training assistance system. The system includes:

[0013] Electromyogram signal processing module: used to obtain the electromyogram signal and posture signal of the user during the strength training process, identify the starting point of the action, and determine the strength training information according to the extracted electromyogram signal characteristics and posture signal characteristics, where the strength training information includes the action type and the muscle group exertion intensity;

[0014] Electroencephalogram signal processing module: obtain the electroencephalogram signal of the user during the strength training process, extract the frequency domain signal characteristics according to the identified starting point of the action, and further determine the brain attention intensity information according to the frequency domain signal characteristics;

[0015] Feedback module: used to analyze in combination with the actual training strength based on the strength training information and the brain attention emphasis information, and feedback the strength training result to the user.

[0016] Compared with the prior art, the advantages of the present invention are as follows: innovatively applying the brain waves for detecting brain activities to strength training, using a portable brain wave and electromyogram acquisition terminal to achieve real-time limb movement feedback and provide assistance for mental concentration, and helping users to perform more efficient strength training. Moreover, the present invention also adopts a data processing algorithm based on an attention mechanism to effectively extract the brain wave signals in the movement cycle, removing the brain wave data irrelevant to the muscle movement time in the data and further amplifying the signal characteristics. In addition, the present invention also uses a long short-term memory network (LSTM) to process the brain wave signals, classifying high mental states and low mental states, significantly improving the accuracy of mental state classification, making it possible to give strength training feedback based on the degree of attention shown by mental activities, and improving the accuracy of strength perception.

[0017] Other features and advantages of the present invention will become clear from the following detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings incorporated in and constituting a part of this specification illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0019] Figure 1 is a flowchart of a multi-modal strength training assistance method according to an embodiment of the present invention;

[0020] Figure 2 is a schematic structural diagram of an acquisition terminal system according to an embodiment of the present invention;

[0021] Figure 3 is a schematic diagram of the working process of the system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Now, various exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and values set forth in these embodiments do not limit the scope of the present invention.

[0023] The following description of at least one exemplary embodiment is merely illustrative in nature and in no way limits the present invention, its application or use.

[0024] Techniques, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods and devices should be regarded as part of the specification.

[0025] In all the examples shown and discussed here, any specific values should be construed as merely exemplary and not as a limitation. Thus, other examples of the exemplary embodiments may have different values.

[0026] It should be noted that like reference numerals and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

[0027] What the present invention provides is a power training assistance method based on multi-modal (electromyogram, limb posture, electroencephalogram). It uses electromyogram signals to judge the intensity of different muscle groups exerting force in different movements, and at the same time detects the mental intensity to judge whether the user is focused on the training itself at this time, and enables the user to immerse in the training through visual feedback. This is an innovative training assistance method. In addition, the attention mechanism is used to extract and amplify the electroencephalogram data of the motion state and classify it through the classification method of deep learning, so as to judge the level of the user's mental intensity. The present invention is a novel power training method and system, that is, it jointly feeds back to the user through the physical state and the brain attention intensity during training to improve the training efficiency of the user.

[0028] In short, the provided multi-modal-based power training assistance method includes the following steps:

[0029] Step S1, obtain the electromyogram signal of the user during the power training process and identify the action starting point, and determine the power training information according to the extracted electromyogram signal features, where the power training information includes the action type and the corresponding muscle group exertion intensity;

[0030] Step S2, obtain the electroencephalogram signal of the user during the power training process, extract the frequency domain signal features according to the identified action starting point, and further determine the brain attention intensity information according to the frequency domain signal features;

[0031] Step S3, based on the power training information and the brain attention emphasis information, combine with the actual training power for analysis to feedback the power training result to the user.

[0032] Specifically, as Figure 1 shown, in one embodiment, step S1 includes the following sub-steps:

[0033] Step S1.1, embed the electromyogram sensor and the nine-axis sensor in the fitness T-shirt to obtain the electromyogram signal and the posture information of specific muscle groups during the power training process.

[0034] Step S1.2, perform short-period windowing and band-pass filtering processing on the continuously sampled electromyogram signal and nine-axis sensor signal.

[0035] Step S1.3: Calculate the median frequency of the EMG signals within a short period in Step S1.2.

[0036] Step S1.4: Determine whether the obtained median frequency of the short - period EMG signals exceeds a set threshold. If it exceeds the set threshold, it indicates that this action has started; if it does not exceed, return to the next window of Step S1.2.

[0037] Step S1.5: Extract features from multi - channel EMG signals and attitude signals.

[0038] Step S1.6: Use a support vector machine to classify the extracted EMG features and attitude features.

[0039] Preferably, in Steps S1.2, S1.3, and S1.4, in the continuously collected multi - channel EMG signals and nine - axis sensor signals, a fixed time duration t (for example, t is set to be less than 0.25 s) is taken as a short - period window, and the signals of this window are filtered for noise through a band - pass filter. Calculate the median frequency of the EMG signals in multiple windows of the multi - channel EMG signals, and compare whether this median frequency exceeds the median frequency of EMG in a general static or non - muscle - exerting state. If it exceeds the threshold, it is considered that the strength training has started. Similarly, if the median frequencies of the multi - channel EMG signals measured at multiple t moments are all less than the threshold, it is considered that this strength training has ended.

[0040] Preferably, in Steps S1.5 and S1.6, mainly calculate the limb mutual angle, median frequency, zero - crossing number, root - mean - square value, waveform length, and average power of the attitude information and multi - channel EMG signals during a single movement as feature values. Pre - train a support vector machine with a Gaussian kernel function for action classification to obtain a trained model. During actual use, use the trained model for action classification.

[0041] In summary, in Step S1, through the method of adaptive cutting, the starting point of a single training is segmented, avoiding artificial recording of the start of the movement. Since the EMG sensor can record the intensity of muscle exertion and the nine - axis attitude sensor can obtain the angles of the corresponding muscle groups, the type of movement can be judged by extracting their features, and then whether the force - exerting method of this movement is correct and whether the angles of the limbs are correct can be judged. At the same time, the three - dimensional model of the human body can be modeled using the obtained attitude and muscle group force information as the data for visual feedback.

[0042] For Step S2, in one embodiment, an innovative method for extracting brain - wave data is adopted, that is, an extraction method using an attention mechanism. The training state of the user at this time is judged through EMG signals to extract the brain - wave signals within the movement cycle to the greatest extent. Still combined with Figure 1 As shown, Step S2 includes the following sub - steps:

[0043] Step S2.1: By embedding an electroencephalogram sensor in a headband, obtain the real-time electroencephalogram signal of the frontal lobe Fp1.

[0044] Step S2.2: Judge the result of whether the strength training starts in Step S1.4. If the result is true, record the electroencephalogram data at this time.

[0045] Step S2.3: Obtain the signal within 1 - 30 Hz of the electroencephalogram through a band-pass filter, and divide the electroencephalogram signal within the starting time period of this complete strength training into n (for example, take n = 8) time windows.

[0046] Step S2.4: Use Fourier transform to obtain the proportion of each integer frequency in each time window within 1 - 30 Hz in the full frequency, and obtain the proportion of each frequency component.

[0047] Step S2.5: Use an attention model to give a larger weight to the data within the time period of high muscle activity during this strength training when inputting into the model. Specifically, denote the proportion of the 1 - 30 Hz frequency components obtained by Fourier transform of the electroencephalogram signal recorded in the i-th time window in Step S2.4 as vector W i =[f1,f2,f3...f 30 , calculate the median frequency of the i-th time window within the measurement time window corresponding to the electromyogram sensor and denote it as MF i , calculate the proportion of MF i in the whole movement and denote it as E i . Multiply the proportion vector W i of each frequency in the i-th time window obtained with the weight E i representing the muscle activity level during the strength training process. That is, the attention model can extract more useful information for subsequent classification by paying more attention to the electroencephalogram signals of the active parts of the muscles during the strength training process.

[0048] Step S2.6: Use the n output vectors obtained in S2.5, and classify the mental effort level using, for example, a long short-term memory network (LSTM).

[0049] Furthermore, Step S3 feeds back to the user through visual display. The user can intuitively feel the degree and angle of muscle exertion during this training, and make adjustments to the exertion method and posture compared with the standard training. At the same time, when the user's attention is not concentrated, remind the user through the perception of mental intensity, and visually display the level of mental intensity to help the user immerse in the strength training and improve the training effect.

[0050] Accordingly, the present invention further provides a multi-modal strength training assistance system for implementing one or more aspects of the above method. For example Figure 2 is a common system composition, where Figure 2 (a) shows the terminals (i.e., EMG sensors and nine-axis attitude sensors) attached to the surface of muscle groups to detect EMG and attitude, Figure 2 (b) shows the head-mounted electroencephalogram sensor and visual feedback module. During strength training, since a group of movements involves multiple muscle groups, in order to improve the exercise efficiency, the purpose of fitness is usually to exercise the force generation of one of the muscle groups. When other muscle groups participate in force generation, it means that the training weight is too heavy for the user. This kind of overloaded training not only reduces the training effect but also often brings the risk of sports injuries. And the present invention specifically proposes a strength training assistance method.

[0051] Specifically, the system includes: a perception and transmission module for real-time collection and transmission of electroencephalogram signals, multi-channel EMG signals, and nine-axis attitude sensor data; a data processing module (such as a microprocessor) for processing the collected multi-modal signals; and a visualization feedback module for giving the user feedback data for correct training actions and immersive training. For example, it helps the user adjust the force generation method in a visual (such as virtual reality) way, informs the user to increase or decrease the weight of the equipment, and helps the user improve the mental focus during the training process.

[0052] In one embodiment, the perception and transmission module includes the following units:

[0053] An acquisition unit for acquiring the EMG signals of specific muscle groups, nine-axis sensor data, and electroencephalogram signals of Fp1 on the forehead;

[0054] An extraction unit for framing and encoding the data acquired by the acquisition unit;

[0055] A transmission unit for real-time transmitting the data packets obtained by the extraction unit to the data processing module via low-power Bluetooth.

[0056] In one embodiment, the data processing module includes the following units:

[0057] A data reception unit for discriminating the data packets received via Bluetooth and determining whether the collected data comes from the EMG signals of a certain channel, nine-axis sensor data, or electroencephalogram signals of Fp1 on the forehead;

[0058] A data processing unit for preprocessing the received EMG signals, feature extraction, adaptive segmentation, band-pass filtering, and preprocessing the received electroencephalogram data, band-pass filtering, adaptive segmentation, and Fourier analysis;

[0059] Modeling unit: For EMG signals, the features extracted from, for example, five different actions in the training set are input into a support vector machine model with a Gaussian kernel function to obtain a pre-trained support vector machine model. And for EEG signals, the high mental attention and low mental attention states in the training set are used as the classification results of a long short-term memory network (LSTM) to train the network, obtaining a pre-trained long short-term memory network (LSTM) model;

[0060] Detection unit: The processed EMG data is input into the support vector machine model, and the EEG data applied with the attention model mechanism is input into the pre-trained long short-term memory network (LSTM) to classify the mental attention level.

[0061] Figure 3 This is the main process of the multi-modal strength training assistance system, including: extracting EMG signals and adaptively cutting the motion window through the average frequency, using the attention mechanism to extract the signals in the EEG signals when the muscle force is strong, and classifying the high mental intensity and low mental intensity through the long short-term memory network (LSTM), providing visual feedback on the current body posture and force intensity of the user, and helping the user better focus on the training through the results of mental recognition.

[0062] It should be noted that without departing from the spirit and scope of the present invention, those skilled in the art can make appropriate changes or modifications to the above embodiments. For example, in addition to the support vector machine, a neural network model can also be used to determine the action type and muscle group force intensity, or GRU (gated recurrent unit) can be used instead of LSTM.

[0063] In summary, the multi-modal strength training assistance method and system provided by the present invention mainly include: using EMG sensors embedded in the T-shirt to measure the EMG signals of the user during strength training, identifying the start of the action, the type of the corresponding action, and the force intensity of the corresponding muscle group; using EEG sensors embedded in the headband to measure the EEG signals of the user during strength training, and obtaining the brain attention intensity within a complete strength training cycle through model classification; using the obtained strength training information and brain attention intensity information, analyzing in combination with the actual training volume of the user, and providing feedback through the visual module to help the user train correctly and efficiently. The present invention innovatively uses EEG signals in combination with EMG signals to assist the user in strength training, which can not only correct the user's posture and force application method through EMG sensors, but also judge the mental intensity of the user during exercise based on the EEG signals of the user's motion state, and use visual feedback to help the user be more immersed in the training.

[0064] The present invention may be a system, a method, and / or a computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions thereon for causing a processor to implement aspects of the present invention.

[0065] A computer-readable storage medium may be a tangible device that can retain and store instructions for use by an instruction execution device. A computer-readable storage medium may be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanically encoded device such as a punch card or raised structures in grooves having instructions stored thereon, and any suitable combination of the foregoing. The computer-readable storage medium as used herein is not to be construed as a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0066] The computer-readable program instructions described herein may be downloaded to respective computing / processing devices from a computer-readable storage medium or may be downloaded to an external computer or external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0067] The computer program instructions for performing the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine - related instructions, microcode, firmware instructions, state - setting data, or source code or object code written in any combination of one or more programming languages, including object - oriented programming languages such as Smalltalk, C++, Python, etc., and conventional procedural programming languages such as the "C" language or similar programming languages. The computer - readable program instructions may be executed entirely on the user's computer, partially on the user's computer, executed as a stand - alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider). In some embodiments, by using the state information of the computer - readable program instructions to customize an electronic circuit, such as a programmable logic circuit, a field - programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit can execute the computer - readable program instructions to implement various aspects of the present invention.

[0068] Aspects of the present invention are described herein with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer - readable program instructions.

[0069] These computer - readable program instructions can be provided to a processor of a general - purpose computer, a special - purpose computer, or other programmable data - processing apparatus to produce a machine such that, when the instructions are executed by the processor of the computer or other programmable data - processing apparatus, a device is produced that implements the functions / acts specified in one or more blocks of the flowchart and / or block diagram. These computer - readable program instructions can also be stored in a computer - readable storage medium, which causes a computer, a programmable data - processing apparatus, and / or other devices to operate in a particular manner. Thus, the computer - readable medium storing the instructions includes a manufacture, which includes instructions for implementing various aspects of the functions / acts specified in one or more blocks of the flowchart and / or block diagram.

[0070] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices, causing a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other devices to produce a computer-implemented process such that the instructions executed on the computer, other programmable data processing apparatus, or other devices implement the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0071] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two consecutive blocks may in fact be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functionality involved. It should also be noted that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are equivalent.

[0072] The embodiments of the present invention have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, the practical application, or the improvement of technologies in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. A multi-modal strength training assistance method, comprising the following steps: Obtain the electromyogram signal and posture signal of the user during strength training, identify the starting point of the action, and determine the strength training information according to the extracted electromyogram signal features and posture signal features, where the strength training information includes the action type and the muscle group force intensity; Obtain the brain wave signal of the user during strength training, extract the frequency domain signal features according to the identified starting point of the action, and further determine the brain attention intensity information according to the frequency domain signal features; Based on the strength training information and the brain attention emphasis information, analyze in combination with the actual training strength, and feedback the strength training result to the user.

2. The method according to claim 1, characterized in that, The identification of the starting point of the action includes: Use the electromyogram sensor and nine-axis sensor embedded in the fitness T-shirt to obtain the electromyogram signal and posture signal of the target muscle group during strength training; Perform short-period windowing and band-pass filtering on the continuously sampled electromyogram signal and posture signal; Calculate the median frequency of the electromyogram signal within a short period; Judge whether the median frequency of the obtained short-period electromyogram signal exceeds the set threshold. If it exceeds the set threshold, it indicates that the action has started. If the median frequencies of the multi-channel electromyogram signals measured at multiple moments are all less than the threshold, it is considered that this strength training is over, where the threshold is determined according to the electromyogram median frequency in the static or non-muscle force state.

3. The method according to claim 2, wherein The short-period windowing and band-pass filtering of the continuously sampled electromyogram signal and posture signal includes: In the continuously collected multi-channel electromyogram signal and posture signal, take a fixed duration t as a short-period window, and the signal of this window is filtered by a band-pass filter to remove noise, where t is set to be less than 0.25 s.

4. The method according to claim 1, characterized in that, The strength training information is obtained according to the following steps: Use the brain wave sensor embedded in the headband to obtain the real-time brain wave signal of the forehead Fp1; Record the brain wave data during the entire strength training process; Obtain the signal within 1 - 30 Hz of the brain wave through a band-pass filter, and divide the brain wave signal in the starting time period of a complete strength training into n time windows; Use Fourier transform to obtain the full-frequency proportion of each integer frequency on each time window within 1 - 30 Hz; Use the attention model to extract the proportion vectors of the frequency components within n time windows, and give a larger weight to the data in the time period with a high muscle activity level during this strength training; Use the obtained n output vectors, and use a long short-term memory network to classify the brain attention intensity information.

5. The method according to claim 4, characterized in that, The use of the attention model to extract the proportion vectors of the frequency components within n time windows and give a larger weight to the data in the time period with a high muscle activity level during this strength training includes: The proportion of the 1-30 Hz frequency component obtained by Fourier-transforming the electroencephalogram signal recorded in the i-th time window is denoted as vector W i =[f1, f2, f3... f 30 . The median frequency is calculated for the i-th time window within the measurement time window of the collected electromyogram signal and denoted as MF i , and the proportion of MF i in the entire movement is calculated and denoted as E i ; Multiply each frequency proportion vector W within the obtained i-th time window i by the weight E representing muscle activity during strength training i .

6. The method according to claim 1, characterized in that Based on the strength training information and the brain attention emphasis information, analyzing in combination with the actual training strength to feedback the strength training result to the user includes the following sub-steps: Analyze the electromyogram signal of the specific muscle group obtained, the user's actual training weight, training action, and the result of brain power classification, correct the user's wrong force application method, and provide a recommended weight; In a visual feedback manner, the action state of the user, the exertion degree of the set muscle groups, and the brain attention degree are fed back.

7. The method according to claim 1, characterized in that, The support vector machine model is used to determine the strength training information, and the input of the support vector machine model is the multi-channel electromyogram signal feature and the posture signal feature.

8. The method according to claim 1, wherein The feedback of the strength training result to the user includes: real-time displaying the body posture through visualization technology, analyzing the limb trajectory, exertion mode, and brain attention intensity information during the movement process, and formulating the training process for the user.

9. A multi-modal strength training assistance system, comprising: An electromyogram signal processing module: configured to acquire the electromyogram signal and the posture signal of the user during the strength training process, identify the action starting point, and determine the strength training information according to the extracted electromyogram signal feature and the posture signal feature, where the strength training information includes the action type and the muscle group exertion intensity; An electroencephalogram signal processing module: acquires the electroencephalogram signal of the user during the strength training process, extracts the frequency domain signal feature according to the identified action starting point, and further determines the brain attention intensity information according to the frequency domain signal feature; A feedback module: configured to analyze based on the strength training information and the brain attention emphasis information, and combine with the real training strength, and feed back the strength training result to the user.

10. A computer-readable storage medium having a computer program stored thereon, wherein, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images