Annular surface electromyography sensor position offset calibration method and device and medium

Through the position offset calibration method of the annular surface electromyography sensor, the polar coordinate system and vector superposition technology are used to solve the electrode offset problem of sparse surface electrode sensors, improving the recognition accuracy and calibration efficiency of electromyography signals, and expanding the application range.

CN118902477AInactive Publication Date: 2024-11-08WUHAN NEURACOM TECH DEV CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202410956546.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing surface electromyography sensors are prone to positional deviations in electrodes during frequent wear or use, resulting in signal changes, affecting action classification and recognition accuracy, especially lack of compatibility with sparse surface electrode sensors, which limits its application range.

Method used

The position offset calibration method of the annular surface electromyography sensor is adopted to construct the channel electrode offset correction matrix through denoising pre-processing, time-domain feature extraction, polar coordinate projection and vector superposition to realize the correction of electrode positions, which is suitable for sparse surface electrodes. sensor.

Benefits of technology

This method can accurately estimate the electrode offset angle, improve the identification and classification accuracy of electromyography signals, and is suitable for high-density and sparse types of electromyography signal sensors, enhancing calibration efficiency and practicality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118902477A_ABST
    Figure CN118902477A_ABST
Patent Text Reader

Abstract

The invention provides a position offset calibration method for an annular surface electromyographic sensor. According to the position offset calibration method, a polar coordinate system which is static relative to the annular surface electromyographic sensor is established. In the polar coordinate system, the change of the surface electromyogram signal can be effectively tracked by monitoring the change of the active polar angle under the polar coordinate system. When the annular surface electromyographic sensor rotates and deviates, the transverse rotation deviation angle of the annular surface electromyographic sensor can be accurately estimated only by comparing a small amount of rotated electromyographic signal data with initial or standard position data, and the method is compatible with high-density array type electromyographic signal sensors, sparse electromyographic signal sensors and the like. And the applicability is high. Besides, according to the position offset calibration method, mapping from the offset sample space to the initial sample space can be completed, offset can be estimated and corrected without depending on a pre-trained complex model, and therefore accurate recognition and classification of the electromyographic signals are guaranteed, and calibration efficiency and practicability are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of biological signal analysis, and more specifically, to a method, device and medium for calibrating the position offset of an annular surface electromyography sensor. Background Art

[0002] Surface electromyography (sEMG) is a muscle electrical activity signal collected directly from the human epidermis, which can reflect the behavior and movement information of the limbs. Due to the non-invasive collection method, wide distribution and relatively stable signal characteristics of sEMG, it has become an indispensable source of information in intelligent prostheses, exoskeletons and other human-machine interaction systems.

[0003] However, surface electromyography sensors have certain limitations in actual use. Especially when worn or used frequently, the electrodes of surface electromyography sensors are prone to positional displacement, which will cause changes in the collected signals, thereby affecting the accuracy of action classification and recognition based on these signals. Therefore, the electrode displacement problem has become a technical bottleneck that limits the widespread application of surface electromyography technology.

[0004] One of the existing surface electromyography sensor offset calibration techniques is to use improved grayscale co-occurrence matrix texture features combined with high-density surface electromyography features. This technology has obvious spatial position independence. Compared with traditional time domain and frequency domain features, this method has achieved significant improvement in the accuracy of action recognition. However, this method is mainly applicable to high-density array electromyography signal sensors, and its compatibility with widely used sparse surface electrode sensors has not been fully verified, which limits its scope of application. Summary of the invention

[0005] In response to at least one defect or improvement need in the prior art, the present application provides a method, device and medium for calibrating the position offset of a ring-shaped surface electromyography sensor, which is at least applicable to the offset calibration of sparse surface electrode sensors, thereby expanding the application scope of the offset calibration method to adapt to sensors.

[0006] To achieve the above objectives, in a first aspect, the present application provides a method for calibrating the position offset of a ring-shaped surface electromyography sensor, comprising:

[0007] Performing denoising preprocessing on the surface electromyographic signals obtained by the electrodes of each channel of the annular surface electromyographic sensor during the first action;

[0008] Perform time domain feature extraction on the surface electromyography signal after denoising preprocessing to obtain the corresponding feature value of each channel;

[0009] The corresponding eigenvalues ​​of each channel are projected into the polar coordinate system as the polar diameter of the polar angle corresponding to each channel;

[0010] Perform vector superposition processing on the vectors corresponding to the eigenvalues ​​of the respective channels in the polar coordinate system to obtain the active polar angle of the superimposed total vector;

[0011] Comparing the active polar angle with the initial placement angle of the annular surface electromyography sensor to obtain a channel electrode offset;

[0012] Constructing a channel electrode offset correction matrix according to the relationship between the mapping transformation of the corresponding eigenvalues ​​of each channel and the channel electrode offset;

[0013] Based on the channel electrode offset correction matrix, correcting the eigenvalue corresponding to the surface electromyography signal obtained at the current electrode position to obtain the corrected eigenvalue;

[0014] The annular surface electromyography sensor comprises a plurality of electrodes evenly distributed along the circumference, and each electrode for collecting the surface electromyography signal is a channel.

[0015] Furthermore, the formula for constructing the channel electrode offset correction matrix according to the relationship between the mapping transformation of the corresponding eigenvalues ​​of each channel and the channel electrode offset includes:

[0016]

[0017] Wherein, Δα represents the channel electrode offset; f(Δα) represents the channel offset relationship value; e represents the number of electrodes uniformly distributed along the circumference of the annular surface electromyography sensor; T e×e (Δα) represents the channel electrode offset correction matrix.

[0018] Furthermore, based on the channel electrode offset correction matrix, the eigenvalue corresponding to the surface electromyography signal obtained at the current electrode position is corrected, and the formula for obtaining the corrected eigenvalue includes:

[0019]

[0020] Wherein, X represents the characteristic value corresponding to the surface electromyography signal obtained at the current electrode position; represents the corrected eigenvalue.

[0021] Furthermore, the time domain feature extraction of the surface electromyographic signal after denoising preprocessing is performed to obtain the formula of the corresponding characteristic value of each channel includes:

[0022]

[0023] Where l represents the length of the time window; fs represents the sampling frequency of each channel electrode of the annular surface electromyography sensor; x i represents the amplitude of the surface electromyography signal collected at the i-th sampling point in the time window; f RMS (x) represents the root mean square eigenvalue of a single time window in a single channel; RMS is used as the feature extraction function, and RMS represents root mean square.

[0024] Furthermore, projecting the eigenvalues ​​corresponding to each channel as the polar diameter of the polar angle corresponding to each channel into the polar coordinate system includes:

[0025] The direction of the first channel is the polar axis reference direction of the polar coordinate system, and the second channel to the eth channel correspond to the polar axis reference direction of the polar coordinate system. The polar angle direction of

[0026] The feature extraction method using the root mean square as the feature extraction function takes the root mean square eigenvalue of the surface electromyographic signal in each time window in the e channels as the polar diameter on each corresponding polar angle; for any sample point from the e channel electrodes, there is:

[0027] X=(p n,1 ,p n,2 ,...,p n,m );

[0028] Among them, p n,m represents the RMS eigenvalue from the nth time window and the mth channel, where m = 1, 2, 3, ..., e; The polar coordinate points are plotted in the polar coordinate system, and the magnitude of each root mean square eigenvalue is used as the polar diameter of the scattered points.

[0029] Furthermore, the vectors corresponding to the eigenvalues ​​of the channels in the polar coordinate system are subjected to vector superposition processing to obtain the active polar angle of the superimposed total vector, including:

[0030] The aforementioned polar coordinate points are taken as vectors to perform vector superposition operation, that is, any scattered point in the polar coordinate system is taken as a vector, and the mapping relationship from the polar coordinate system to the rectangular coordinate system can be expressed as p in the rectangular coordinate system. n,m ;

[0031]

[0032] The total superposition vector of the first action is obtained by using vector superposition operation:

[0033]

[0034]

[0035] Wherein, h represents the number of myoelectric feature samples obtained under the guidance of the first action; represents the vector point coordinates of the superimposed total vector; p x p represents the horizontal coordinate of the vector point of the superimposed total vector; y represents the ordinate of the vector point of the superimposed total vector; α represents the active polar angle relative to the superimposed total vector.

[0036] Furthermore, the step of comparing the active polar angle with the initial placement angle of the annular surface electromyography sensor to obtain the channel electrode offset comprises:

[0037] The channel electrode offset is obtained by subtracting the active polar angle from the initial placement angle of the annular surface electromyography sensor.

[0038] Furthermore, the denoising preprocessing includes:

[0039] One or more denoising preprocessing methods including baseline drift removal, power frequency interference removal, high-pass filtering, full-wave rectification, low-pass filtering and normalization.

[0040] Furthermore, the annular surface electromyography sensor includes e electrodes uniformly distributed along the circumference, wherein 3≤e≤16.

[0041] Furthermore, e=8.

[0042] In a second aspect, the present application provides an electronic device comprising at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit is enabled to perform the steps of the position offset calibration method described in any one of the above items.

[0043] In a third aspect, the present application provides a storage medium storing a computer program executable by an access authentication device, wherein when the computer program runs on the access authentication device, the access authentication device is enabled to perform the steps of any of the above-mentioned position offset calibration methods.

[0044] In general, the above technical solutions conceived by the present application can achieve the following beneficial effects compared with the prior art:

[0045] The present application proposes a position offset calibration method for an annular surface electromyographic sensor, which establishes a polar coordinate system that is stationary relative to the annular surface electromyographic sensor. In this polar coordinate system, the changes in the surface electromyographic signal can be effectively tracked by monitoring the changes in the active polar angle under the polar coordinate system. When the annular surface electromyographic sensor undergoes rotational offset, only a small amount of the electromyographic signal data after rotation is compared with the data of the initial or standard position to accurately estimate the lateral rotation offset angle of the annular surface electromyographic sensor. This method is compatible with high-density array, sparse and other types of electromyographic signal sensors, and has strong applicability. In addition, the position offset calibration method can complete the mapping of the sample space after offset to the initial sample space, and the offset can be estimated and corrected without relying on a pre-trained complex model, thereby ensuring the accurate recognition and classification of the electromyographic signal and improving the calibration efficiency and practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of a method for calibrating position offset of a ring-shaped surface electromyography sensor provided in an embodiment of the present application;

[0048] Figure 2 A schematic diagram of a cross section of a ring-shaped surface electromyography sensor provided in an embodiment of the present application and its distribution in a polar coordinate system;

[0049] Figure 3 A flowchart of denoising preprocessing provided in an embodiment of the present application;

[0050] Figure 4 A schematic diagram of a sliding window for surface electromyography signals provided in an embodiment of the present application;

[0051] Figure 5 A schematic diagram of the clustered distribution of the eigenvalue scatter points corresponding to each channel and the active polar angle of the related superimposed total vector provided in the embodiment of the present application;

[0052] Figure 6 A schematic diagram of the lateral rotational offset of the annular surface electromyography sensor provided in an embodiment of the present application;

[0053] Figure 7 A mapping relationship diagram of each channel offset at a special angle of m×45° provided in an embodiment of the present application;

[0054] Figure 8A block diagram of an electronic device provided in an embodiment of the present application and suitable for implementing the position offset calibration method described above. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0056] The terms "first", "second" or "e" in the specification, claims or the above-mentioned drawings of this application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" or "having" 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 limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0057] As described in the background technology section of the specification, one method of the existing surface electromyography sensor offset calibration technology is to use improved grayscale co-occurrence matrix texture features combined with high-density surface electromyography features, and this technology has obvious spatial position independence. Compared with traditional time domain and frequency domain features, this method has achieved significant improvement in the accuracy of action recognition. However, this method is mainly applicable to high-density array electromyography signal sensors, and its compatibility with widely used sparse surface electrode sensors has not been fully verified, which limits its scope of application. In view of this technical status, the present application proposes a ring-shaped surface electromyography sensor position offset calibration method, equipment and medium, which are used for at least offset calibration of sparse surface electrode sensors, thereby expanding the application scope of sensors adapted to the offset calibration method.

[0058] refer to Figure 1 An embodiment of the present application proposes a method for calibrating the position offset of a ring-shaped surface electromyography sensor. The method for calibrating the position offset may include the following 7 steps.

[0059] In one embodiment, Figure 2The annular surface electromyography sensor on the left includes 8 electrodes evenly distributed along the circumference, and each electrode that collects the surface electromyography signal is a channel, so it is an annular surface electromyography sensor with 8 channels. In fact, the number of electrodes e evenly distributed along the circumference of the annular surface electromyography sensor can be taken as 3≤e≤16. Within this value range, the distribution of the surface electromyography sensor is relatively sparse. The annular surface electromyography sensor position offset calibration method proposed in this application can be adapted to the application scenario where the distribution of this surface electromyography sensor is relatively sparse. When the sensors (i.e., electrodes) are distributed more densely, multiple sensors may be attached to the same muscle, resulting in little difference, which will increase the difficulty of algorithm correction; and when the sensors are distributed more sparsely, although the difficulty of algorithm correction will be reduced, the interval between sensors will gradually increase, thereby reducing the necessity of offset correction. This application has been verified by a large number of experiments. When e is 8, that is, when the annular surface electromyography sensor only includes 8 electrodes evenly distributed along the circumference, the best balance between the convenience and necessity of correction can be achieved, thereby maximizing the total benefit of correction. In one embodiment of the present application, the scheme is explained by taking e as 8 as an example. The same is true when e takes other values, and no further details are given.

[0060] Figure 2 The right figure shows the distribution of the cross section of the annular surface electromyography sensor in the polar coordinate system, that is, channel 1 corresponds to a polar angle direction of 0°, channel 2 corresponds to a polar angle direction of 45°..., and channel 8 corresponds to a polar angle direction of 315°.

[0061] The general idea of ​​an embodiment of the position offset calibration method of the present application is: before the offset correction is performed, in order to ensure that the APA (Activation Polar Angle) angle is a constant value, it is necessary to ensure that it stays in a fixed action for 3-5 seconds, that is, the user needs to maintain a fixed action (such as turning the palm up or pressing down and other body movements to remain stationary for a period of time) for 3-5 seconds during the electrode offset correction process, and the electrode collects data during this period. For example, different gestures use different muscle areas, which makes the signal strength collected by the electrode different, which will cause the active polar angle to change and affect the correction. Therefore, it is necessary to maintain a fixed action for 3-5 seconds to keep the active polar angle relatively constant. Then, the original collected surface electromyographic signal is preprocessed to remove interference signals such as noise; the root mean square (RMS) of the preprocessed data is extracted and projected into the polar coordinate system; the active polar angle is found through these eigenvalues, and the difference with the initial position is calculated; finally, the corrected root mean square eigenvalue is output through the correction matrix for subsequent analysis.

[0062] Step 1: Perform denoising preprocessing on the surface electromyographic signals obtained by each channel electrode of the annular surface electromyographic sensor during the first action.

[0063] Since the original surface electromyography signal is interfered by noise, the surface electromyography signal is relatively weak and it is impossible to extract the effective signal related to the first action. The original surface electromyography signal needs to go through a series of processing before it can be used for subsequent analysis. The original surface electromyography signal is processed by removing baseline drift, power frequency interference, high-pass filtering, full-wave rectification, low-pass filtering and normalization to obtain a relatively pure surface electromyography signal. For the specific denoising preprocessing process, please refer to Figure 3 In some embodiments, step 1 may specifically include the following sub-steps.

[0064] Step 11: Remove baseline drift. Baseline drift is caused by inherent electrode noise, motion artifacts or physiological artifacts. Low-frequency noise (0-1Hz) is mixed into the original signal, causing the data time domain waveform to drift in a certain direction and the main frequency part in the frequency domain to be submerged by noise. Methods for removing baseline drift include adaptive filtering, high-pass filtering and sliding average method.

[0065] Step 12: Remove power frequency interference. Power frequency interference is the interference caused by the alternating current that powers electronic equipment in the environment. The frequency of this interference is 50Hz in my country and 60Hz in the United States and some other countries. The method to remove power frequency interference is to use a notch filter to filter out the 50Hz power frequency noise.

[0066] Step 13: High-pass filtering: According to the frequency characteristics of surface electromyographic signals, the effective electromyographic signal frequency is between 5-500Hz, and its energy distribution is effective only between 20Hz-150Hz. Therefore, a high-pass filter with a cutoff frequency of 5Hz is required to suppress signals below 5Hz from passing through.

[0067] Step 14: Full-wave rectification. That is, flip all negative parts in the surface electromyographic signal to the positive domain, so that all activities in each cycle (whether originally positive or negative) are reflected in the form of positive values. Since the amplitude of the surface electromyographic signal is positively correlated with the contraction strength of the muscle, the rectified signal can more intuitively represent the force generated by the muscle at any given time, so that when calculating the total energy of the signal, the contribution of all muscle activities is taken into account, and the energy is not underestimated due to the mutual cancellation of positive and negative waveforms in the original signal. Therefore, this can more accurately reflect the total energy of muscle activity for subsequent feature extraction and pattern recognition.

[0068] Step 15: Low-pass filtering. Since the effective EMG signal frequency is between 5-500Hz, a low-pass filter with a cutoff frequency of 500Hz is required to suppress the signal above 500Hz. In addition, in order to more clearly display the trend and changes of muscle activity, the original signal needs to be smoothed to generate the waveform after envelope extraction.

[0069] Step 16: Normalize. The value of the surface electromyography signal is distributed between 0 and 1 for subsequent processing.

[0070] Step 2: Extract the time domain features of the surface electromyography signal after denoising preprocessing to obtain the corresponding eigenvalues ​​of each channel.

[0071] In some embodiments, more specifically, step 2 is to extract time domain features based on the pure surface electromyographic signal obtained after denoising preprocessing in step 1, and the feature extraction method used is root mean square (RMS), the size of which can be used to represent the contribution of each muscle group during exercise. The fragment data at the moment of muscle movement is intercepted, and the root mean square is used as the feature extraction function to construct a feature space constructed by the root mean square features of each channel.

[0072] Figure 4 A schematic diagram of a surface electromyography signal sliding window provided in an embodiment of the present application, wherein the time window length l is set to 400ms, the incremental window length k is set to 200ms, and sliding is performed in the timing waveform. The root mean square eigenvalue f of a single time window in a single channel RMS The expression of (x) is as follows:

[0073]

[0074] Among them, x i is the amplitude of the surface electromyography signal collected at the i-th sampling point in the time window; f s is the sampling frequency of each channel electrode of the annular surface electromyography sensor.

[0075] Step 3: Project the corresponding eigenvalues ​​of each channel into the polar coordinate system as the polar diameter of the polar angle corresponding to each channel.

[0076] In some embodiments, more specifically, the direction of channel 1 is taken as the polar axis reference direction of the polar coordinate system, and channels 2, 3, 4, ..., 8 correspond to the polar angle directions of 45°, 90°, 135°, ..., 315° in the polar coordinate system respectively; according to the root mean square feature extraction method in step 2, the root mean square eigenvalue of the surface electromyographic signal in each time window in the 8 channels is used as the polar diameter at each polar angle, and the drawing and visualization principle of the scatter plot in the polar coordinate system is as follows: for any sample point from the 8-channel sensor:

[0077] X=(p n,1 ,p n,2 ,...,p n,m );

[0078] Among them, p n,mrepresents the RMS eigenvalue of the mth channel from the nth time window, where m = 1, 2, 3, ..., 8. n,m ,m×45°) are plotted in the polar coordinate system in the form of polar coordinate points, that is, the RMS eigenvalue of channel 1 is plotted on the 0° polar axis, the RMS eigenvalue of channel 2 is plotted on the 45° polar axis, and so on, the RMS eigenvalue of channel 8 is plotted on the 315° polar axis, where the size of each RMS eigenvalue is used as the polar diameter of the scattered point drawing, and finally forms Figure 5 The black scattered points are concentrated in the distribution area.

[0079] Step 4: Perform vector superposition processing on the vectors corresponding to the eigenvalues ​​of each channel in the polar coordinate system to obtain the active polar angle of the superimposed total vector.

[0080] In some embodiments, more specifically, the polar coordinate points obtained in step 3 are used as vectors for vector superposition operation, that is, any scattered points in the polar coordinate system are used as vectors, and the mapping relationship from the polar coordinate system to the rectangular coordinate system can be expressed as follows in the rectangular coordinate system:

[0081]

[0082] The overall resultant vector under this action is calculated by vector superposition, and the polar angle of the overall resultant vector is defined as the active polar angle α. The calculation principle is:

[0083]

[0084] Among them, h represents the sample size of the electromyographic features obtained under the guidance of a certain action label, so that the active polar angle α in the polar coordinates of the action can be obtained, such as Figure 5 Active polar angle APA shown.

[0085] Step 5: Compare the active polar angle with the initial placement angle of the annular surface electromyography sensor to obtain the channel electrode offset.

[0086] In some embodiments, more specifically, the channel electrode offset Δα is obtained by subtracting the active polar angle α calculated in step 4 from the initial placement angle of the worn annular surface electromyography sensor. Assume that all channels are offset by Δα (Δα∈[0°,360°)) relative to their corresponding initial positions, such as Figure 6 As shown in Figure 1, when Δα∈[0°,45°), the post-migration channel 3 mainly comes from the pre-migration channel 3 and the pre-migration channel 4. The superimposed signal can be obtained by weighting the two, and the weight factor is negatively correlated with the deflection angle of the two. Therefore, when Δα∈[0°,45°), the characteristic signal after calibration is It can be expressed as:

[0087]

[0088] Wherein, the definitions of n and m are the same as described above. When Δα is taken as a special offset angle of exactly m×45°, where m=1,2,3,...,8, the channel 3 electrode after the offset is exactly located at the muscle position where channel m is located in the initial position, and the measurement positions of other offset channels are the same.

[0089] Therefore, the relationship between the RMS eigenvalue of the channel after migration and the channel before migration can be constructed. When the migration angle is between [0°, 45°) and [315°, 360°), it is related to the RMS eigenvalue of the channel before migration, such as Figure 7 As shown, the mapping relationship of each channel offset by a special angle of m×45° can be obtained.

[0090] Step 6: construct a channel electrode offset correction matrix according to the relationship between the mapping transformation of the corresponding eigenvalues ​​of each channel and the channel electrode offset.

[0091] In some embodiments, more specifically, the mapping relationship between the special positions in step 5 is piecewise linearized, and then Δα is expanded to represent any electrode deflection angle within [0,360°); f(Δα) represents the channel offset relationship, and its expression is:

[0092]

[0093] And from this, the mapping relationship between the RMS eigenvalue of each channel and the RMS eigenvalue after offset can be extended to obtain the channel electrode offset correction matrix T 8×8 (Δα), which is expressed as:

[0094]

[0095] Step 7: Based on the channel electrode offset correction matrix, correct the eigenvalues ​​corresponding to the surface electromyography signals obtained at the current electrode position to obtain the corrected eigenvalues.

[0096] In some embodiments, more specifically, the offset data is multiplied by the channel electrode offset correction matrix obtained in step 6 to obtain the corrected root mean square eigenvalue. That is, the offset data is corrected by compensation between adjacent channels, and the formula is as follows:

[0097]

[0098] Therefore, when the electrode position undergoes a lateral rotational offset, the interference electromyographic data X at the current electrode position (the characteristic value corresponding to the surface electromyographic signal obtained at the current electrode position) is obtained, and the offset Δα from the initial position is calculated. Then, the electromyographic features after the offset interference are mapped to the original sample space to obtain the corrected characteristic signal (Corrected eigenvalues).

[0099] The present application proposes a position offset calibration method for an annular surface electromyographic sensor, which establishes a polar coordinate system that is stationary relative to the annular surface electromyographic sensor. In this polar coordinate system, the changes in the surface electromyographic signal can be effectively tracked by monitoring the changes in the active polar angle under the polar coordinate system. When the annular surface electromyographic sensor undergoes rotational offset, only a small amount of the electromyographic signal data after rotation is compared with the data of the initial or standard position to accurately estimate the lateral rotation offset angle of the annular surface electromyographic sensor. This method is compatible with high-density array, sparse and other types of electromyographic signal sensors, and has strong applicability. In addition, the position offset calibration method can complete the mapping of the sample space after offset to the initial sample space, and the offset can be estimated and corrected without relying on a pre-trained complex model, thereby ensuring the accurate recognition and classification of the electromyographic signal and improving the calibration efficiency and practicality.

[0100] Figure 8 A block diagram of an electronic device suitable for implementing the position offset calibration method described above according to an embodiment of the present application is schematically shown. Figure 8 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0101] like Figure 8 As shown, the electronic device 1000 described in this embodiment includes: a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage part 1008 to a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include an onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the position offset calibration method flow according to an embodiment of the present application.

[0102] In RAM 1003, various programs and data required for the operation of system 1000 are stored. Processor 1001, ROM 1002 and RAM 1003 are connected to each other through bus 1004. Processor 1001 performs various operations of the position offset calibration method flow according to the embodiment of the present application by executing the program in ROM 1002 and / or RAM 1003. It should be noted that the program can also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 can also perform various operations of the position offset calibration method flow according to the embodiment of the present application by executing the program stored in the one or more memories.

[0103] According to an embodiment of the present application, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to the bus 1004. The system 1000 may further include one or more of the following components connected to the I / O interface 1005: an input part 1006 including a keyboard, a mouse, etc.; an output part 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage part 1008 including a hard disk, etc.; and a communication part 1009 including a network interface card such as a LAN card, a modem, etc. The communication part 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that the computer program read therefrom is installed into the storage part 1008 as needed.

[0104] The position offset calibration method process according to the embodiment of the present application can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains a program code for executing the position offset calibration method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or installed from the removable medium 1011. When the computer program is executed by the processor 1001, the above-mentioned functions defined in the system of the embodiment of the present application are executed. According to an embodiment of the present application, the systems, devices, means, modules and / or units described above can be implemented by computer program modules.

[0105] The embodiments of the present application also provide a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the steps of the position offset calibration method according to the embodiments of the present application can be implemented.

[0106] According to an embodiment of the present application, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present application, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus, or a device. For example, according to an embodiment of the present application, the computer-readable storage medium may include one or more memories other than the ROM 1002 and / or RAM 1003 described above.

[0107] It should be noted that the functional modules in the various embodiments of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product.

[0108] The flowchart and / or block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method or computer program product according to various embodiments of the present application. In this regard, each box in the flowchart and / or block diagram can represent a part of a module, program segment or code, and a part of the above-mentioned module, program segment or code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. It should also be noted that each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0109] Those skilled in the art will appreciate that the features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present application. In particular, without departing from the spirit and teachings of the present application, the technical features described in the various embodiments and / or claims of the present application may be combined and / or combined in a variety of ways, and all of these combinations and / or combinations fall within the scope of the present application.

[0110] Although the present application has been shown and described with reference to specific exemplary embodiments of the present application, it should be understood by those skilled in the art that various changes in form and details may be made to the present application without departing from the spirit and scope of the present application as defined by the appended claims and their equivalents. Therefore, the scope of the present application should not be limited to the above-mentioned embodiments, but should be determined not only by the appended claims, but also by the equivalents of the appended claims.

Claims

1. A method for calibrating the position offset of a ring-shaped surface electromyography sensor, characterized in that: include: Performing denoising preprocessing on the surface electromyographic signals obtained by the electrodes of each channel of the annular surface electromyographic sensor during the first action; Perform time domain feature extraction on the surface electromyography signal after denoising preprocessing to obtain the corresponding feature value of each channel; The corresponding eigenvalues ​​of each channel are projected into the polar coordinate system as the polar diameter of the polar angle corresponding to each channel; Perform vector superposition processing on the vectors corresponding to the eigenvalues ​​of the respective channels in the polar coordinate system to obtain the active polar angle of the superimposed total vector; Comparing the active polar angle with the initial placement angle of the annular surface electromyography sensor to obtain a channel electrode offset; Constructing a channel electrode offset correction matrix according to the relationship between the mapping transformation of the corresponding eigenvalues ​​of each channel and the channel electrode offset; Based on the channel electrode offset correction matrix, correcting the eigenvalue corresponding to the surface electromyography signal obtained at the current electrode position to obtain the corrected eigenvalue; The annular surface electromyography sensor comprises a plurality of electrodes evenly distributed along the circumference, and each electrode for collecting the surface electromyography signal is a channel.

2. The position offset calibration method according to claim 1, characterized in that: The formula for constructing the channel electrode offset correction matrix according to the relationship between the mapping transformation of the corresponding eigenvalues ​​of each channel and the channel electrode offset includes: Wherein, Δα represents the channel electrode offset; f(Δα) represents the channel offset relationship value; e represents the number of electrodes uniformly distributed along the circumference of the annular surface electromyography sensor; T e×e (Δα) represents the channel electrode offset correction matrix.

3. The position offset calibration method according to claim 2, characterized in that: Based on the channel electrode offset correction matrix, the eigenvalue corresponding to the surface electromyography signal obtained at the current electrode position is corrected, and the formula for obtaining the corrected eigenvalue includes: Wherein, X represents the characteristic value corresponding to the surface electromyography signal obtained at the current electrode position; represents the corrected eigenvalue.

4. The position offset calibration method according to any one of claims 2 to 3, characterized in that: The formula for extracting time domain features of the surface electromyographic signal after denoising preprocessing to obtain the corresponding characteristic values ​​of each channel includes: Where, l represents the length of the time window; f s represents the sampling frequency of each channel electrode of the annular surface electromyography sensor; x i represents the amplitude of the surface electromyography signal collected at the i-th sampling point in the time window; f RMS (x) represents the root mean square eigenvalue of a single time window in a single channel; RMS is used as the feature extraction function, and RMS represents root mean square.

5. The position offset calibration method according to claim 4, characterized in that: The step of projecting the characteristic values ​​corresponding to each channel as the polar diameter of the polar angle corresponding to each channel into the polar coordinate system includes: The direction of the first channel is the polar axis reference direction of the polar coordinate system, and the second channel to the eth channel correspond to the polar axis reference direction of the polar coordinate system. The polar angle direction of The feature extraction method using the root mean square as the feature extraction function takes the root mean square eigenvalue of the surface electromyographic signal in each time window in the e channels as the polar diameter on each corresponding polar angle; for any sample point from the e channel electrodes, there is: X=(p n,1 ,p n,2 ,...,p n,m ); Among them, p n,m represents the RMS eigenvalue from the nth time window and the mth channel, where m = 1, 2, 3, ..., e; The polar coordinate points are plotted in the polar coordinate system, and the magnitude of each root mean square eigenvalue is used as the polar diameter of the scattered points.

6. The position offset calibration method according to claim 5, characterized in that: The process of performing vector superposition processing on the vectors corresponding to the eigenvalues ​​of the respective channels in the polar coordinate system to obtain the active polar angle of the superimposed total vector includes: The aforementioned polar coordinate points are taken as vectors to perform vector superposition operation, that is, any scattered point in the polar coordinate system is taken as a vector, and the mapping relationship from the polar coordinate system to the rectangular coordinate system can be expressed in the rectangular coordinate system as The total superposition vector of the first action is obtained by using vector superposition operation: Wherein, h represents the number of myoelectric feature samples obtained under the guidance of the first action; represents the vector point coordinates of the superimposed total vector; p x p represents the horizontal coordinate of the vector point of the superimposed total vector; y represents the ordinate of the vector point of the superimposed total vector; α represents the active polar angle relative to the superimposed total vector.

7. The position offset calibration method according to claim 6, characterized in that: The step of comparing the active polar angle with the initial placement angle of the annular surface electromyography sensor to obtain the channel electrode offset comprises: The channel electrode offset is obtained by subtracting the active polar angle from the initial placement angle of the annular surface electromyography sensor.

8. The position offset calibration method according to any one of claims 1 to 3 and claims 5 to 7, characterized in that: The annular surface electromyography sensor includes e electrodes uniformly distributed along the circumference, wherein 3≤e≤16.

9. An electronic device, characterized in that: It comprises at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit is enabled to perform the steps of the position offset calibration method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: It stores a computer program executable by an access authentication device. When the computer program runs on the access authentication device, the access authentication device is enabled to execute the steps of the position offset calibration method according to any one of claims 1 to 8.

Citation Information

Patent Citations

  • Surface electromyographic signal gesture recognition method

    CN111367399A