Method for adjusting sampling frequency of electromyographic signals

Abstract

The invention relates to the technical field of data processing, and discloses a method for adjusting the sampling frequency of electromyographic signals. According to the method, electric signal data generated by muscles are collected in real time through installed electrode detection equipment, an electric signal data matrix is constructed, and meanwhile the collected electric signal data matrix is processed in a data processing mode; analyzing the processed electric signal data matrix through a data analysis method; carrying out feature extraction on the analyzed electric signal data through a feature extraction algorithm; and after the extraction is completed, summarizing the electric signal data after the feature extraction to construct an SVM model, training the constructed SVM model through a training optimization method to obtain a final SVM model, carrying out classification processing on the electric signal data after the feature extraction based on the final SVM model, and finally constructing an electric signal data grade system. And the electromyographic signal sampling accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for adjusting the sampling frequency of an electromyographic signal. Background Art

[0002] As surface electromyography (EMG) technology becomes increasingly popular in rehabilitation medicine, sports training, and human-computer interaction, there is a need for surface electromyography (EMG) signal acquisition and processing equipment that is practical, portable, high-quality, low-cost, and easy to use. However, due to the variability of surface electromyography (EMG) signals, signal acquisition is easily affected by many factors.

[0003] The prior art, such as the invention patent application with announcement number: CN114668563B, discloses a multi-level adjustment method for the sampling frequency of an electromyographic signal. The invention obtains a first electromyographic signal, and when it is determined that the target action corresponding to the first electromyographic signal is a continuous action, the electromyographic sampling frequency of the bionic hand is lowered according to the first value; the second electromyographic signal is obtained, and when it is determined that the second electromyographic signal is the target electromyographic signal, the cumulative consecutive number of times the target electromyographic signal is collected is obtained, wherein the second electromyographic signal is the electromyographic signal obtained after the first electromyographic signal, and the target electromyographic signal is one of several high-frequency electromyographic signals; the increase value is determined according to the cumulative consecutive number of times, and the electromyographic sampling frequency is increased according to the increase value.

[0004] As can be seen from the above scheme, the current sampling of electromyographic signals is often performed by frequency modulation after identification, which ignores the difference of electromyographic signals and has certain limitations in the collected data. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a method for adjusting the sampling frequency of electromyographic signals, which has the advantages of accuracy, real-time and high efficiency, and solves the problem that the acquisition of surface electromyographic signals is easily affected by multiple factors due to the differences in surface electromyographic signals.

[0007] (2) Technical solution

[0008] In order to solve the above technical problem that the acquisition of surface electromyographic signals is easily affected by various factors due to the differences in surface electromyographic signals, the present invention provides the following technical solutions:

[0009] This embodiment discloses a method for adjusting the sampling frequency of an electromyographic signal, which specifically includes the following steps:

[0010] S1. Installing a detection electrode device on the muscle surface and collecting electrical signal data generated by the muscle in real time through the installed electrode detection device; and obtaining a matrix of the collected electrical signal data through a matrix construction method;

[0011] S2. Processing the collected electrical signal data matrix by a data processing method to obtain a processed electrical signal data matrix;

[0012] S21, removing noise from the collected electrical signal data matrix by a data noise reduction method to obtain a noise-reduced electrical signal data matrix;

[0013] S22, filtering the noise-reduced electrical signal data by a data filtering method to obtain a filtered electrical signal data matrix;

[0014] S23, setting the filtered electrical signal data matrix as the processed electrical signal data matrix;

[0015] S3. Analyze the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data; and simultaneously extract features from the analyzed electrical signal data using a feature extraction algorithm, and output the feature-extracted electrical signal data;

[0016] S4, summarizing the electrical signal data after feature extraction to construct an SVM model, and training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the electrical signal data after feature extraction based on the final SVM model to obtain classified electrical signal data;

[0017] S5. Construct an electrical signal data hierarchy system based on the characteristics of the electrical signal, and perform frequency modulation management on the classified electrical signal data based on the constructed electrical signal data hierarchy system.

[0018] Preferably, the method of installing a detection electrode device on the muscle surface and collecting electrical signal data generated by the muscle in real time through the installed electrode detection device; and obtaining a matrix of collected electrical signal data through a matrix construction method comprises the following steps:

[0019] S11, each collection point in the installed electrode detection device includes two sets of electrodes and a differential amplifier; the electrical signals collected by the two sets of electrodes are set as a set of electrical signal data generated by the muscles in real time; after the electrical signal data collection is completed, the collected electrical signal data is amplified by the differential amplifier;

[0020] Set the electrode detection equipment collection time point set a={a1,a2,...,a i ,...,a a′}, a i represents the set i-th acquisition time point, and a′ represents the total number of set acquisition time points;

[0021] S12. Summarize the electrical signal data collected at each set collection time point to obtain a first data matrix; and use the first data as the electrical signal data matrix, as shown below:

[0022]

[0023] Among them, y(a ia' ) represents the data collected by the i-th collection point in the electrode detection device at the a'th time point.

[0024] Preferably, the removing noise from the collected electrical signal data matrix by using a data noise reduction method to obtain a noise-reduced electrical signal data matrix comprises the following steps:

[0025] S211, transforming the electrical signal data through a wavelet transform function;

[0026] The wavelet transform function is as follows:

[0027]

[0028] Where m is the scale factor, n is the displacement factor, and f(t) is the square integrable function. is the mother wavelet, W f (m,n) is the wavelet transform function;

[0029] S212: After the transformation is completed, a threshold is set to perform noise reduction processing on the electrical signal data in the electrical signal data matrix;

[0030] The noise expression of the electrical signal data in the electrical signal data matrix is ​​as follows:

[0031] y(t)=f(t)+σ*e(t);

[0032] Where y(t) is the electrical signal data in the electrical signal data matrix, σ is the standard deviation of the noise, and e(t) is the noise;

[0033] The electric signal data after wavelet transformation is divided by the threshold function, and the electric signal data smaller than the threshold is adjusted to 0;

[0034] Threshold function:

[0035]

[0036] Among them, λ is the set threshold, y(t) j,k is the amplitude of the electrical signal data, j represents the horizontal coordinate of the amplitude, k represents the vertical coordinate of the amplitude, sign is the sign function, is the electrical signal data after noise reduction.

[0037] Preferably, filtering the noise-reduced electrical signal data by data filtering to obtain a filtered electrical signal data matrix comprises the following steps:

[0038] Setting the signal-to-noise ratio as a noise reduction evaluation index standard, and filtering the noise-reduced electrical signal data based on the set noise reduction evaluation index standard;

[0039] The signal-to-noise ratio formula is as follows:

[0040]

[0041] Where SNR stands for signal-to-noise ratio, Represents the electrical signal data after noise reduction;

[0042] The higher the signal-to-noise ratio is set, the stronger the noise suppression effect is;

[0043] Set a signal-to-noise ratio threshold. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is less than the set signal-to-noise ratio threshold, it indicates that the noise reduction result is unqualified. The electrical signal data with unqualified noise reduction results are deleted and filtered. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is greater than or equal to the set signal-to-noise ratio threshold, it indicates that the noise reduction result is qualified.

[0044] The filtered electrical signal data are aggregated to form a filtered electrical signal data matrix.

[0045] Preferably, the processing of the electric signal data matrix by a data analysis method to obtain analyzed electric signal data; and extracting features from the analyzed electric signal data by a feature extraction algorithm, and outputting the feature-extracted electric signal data comprises the following steps:

[0046] S31, analyzing the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data;

[0047] S32. After determining the amplitude range of the electrical signal data in the processed electrical signal data matrix, feature extraction is performed on the analyzed electrical signal data using a feature extraction algorithm.

[0048] Preferably, the step of analyzing the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data comprises the following steps:

[0049] Performing modulation analysis on the processed electric signal data matrix using an amplitude keying modulation method to determine the amplitude range of the electric signal data in the processed electric signal data matrix;

[0050] The time domain expression of amplitude keying modulation is:

[0051]

[0052] Among them, the pulse width of the electrical signal d is R d , S MASK (t) is the time domain of amplitude keying modulation, h η is the value of the ηth electrical signal, f k Expressed as the carrier frequency of the kth electrical signal;

[0053] The time domain of the amplitude keying modulation is summed to obtain the electrical signal data after the amplitude keying modulation analysis;

[0054]

[0055] Furthermore, the electrical signal data after amplitude shift keying modulation analysis is summarized to determine the amplitude range of the electrical signal data in the processed electrical signal data matrix.

[0056] Preferably, after determining the amplitude range of the electrical signal data in the processed electrical signal data matrix, simultaneously performing feature extraction on the analyzed electrical signal data using a feature extraction algorithm comprises the following steps:

[0057] S321, dividing the electrical signal data in the processed electrical signal data matrix into K data blocks of uniform size, and using the divided data blocks as input to the convolutional neural network;

[0058] The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer;

[0059] S322: Based on the divided data blocks, set the convolution kernel size and move it on the input image data blocks according to the set step size, and perform convolution calculation on the data in the data blocks in the corresponding area during the movement, so as to achieve feature extraction for each data block;

[0060] The convolution calculation formula is as follows:

[0061] F = y × w + b;

[0062] Among them, y represents the input data block, w represents the weight of the corresponding convolution kernel, b represents the bias value, and F represents the output feature;

[0063] Furthermore, it is assumed that the convolution layer and the pooling layer are continuously stacked, and the output of the upper layer is the input of the lower layer. After all the convolution and pooling operations are completed, the output image data block features are passed to the fully connected layer;

[0064] The output image data block features are expanded and combined through the fully connected layer to obtain a feature data and save it.

[0065] Preferably, the step of aggregating the feature-extracted electrical signal data to construct an SVM model, training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the feature-extracted electrical signal data based on the final SVM model to obtain the classified electrical signal data comprises the following steps:

[0066] S41, constructing an initial SVM model; setting the kernel function used by the initial SVM model to be a Gaussian kernel function; as follows,

[0067] The mathematical expression of the support vector machine algorithm is as follows:

[0068]

[0069] Among them, w represents the weight, b represents the bias value, and φ(c) represents a nonlinear mapping from low-dimensional to high-dimensional space. Represents a sample The classification results;

[0070] in,

[0071] Where, represents the Gaussian kernel function; Represents data in the SVM classification space; represents the core of the Gaussian kernel function; express and The Euclidean distance between them; d represents the range of the Gaussian kernel function;

[0072] S42, set the training data ratio to e; divide the electrical signal data after feature extraction according to the training data ratio e to obtain a training data matrix and a test data matrix; set the maximum number of iterations to e' and the training error threshold to

[0073] The training data matrix is ​​input into the initial SVM model to train the initial SVM model. When the number of training iterations is greater than or equal to e' or the training error is less than When , the training of the initial SVM model is stopped to obtain a trained SVM model;

[0074] S43. Set a test accuracy threshold, input the test data matrix into the trained SVM model for classification, and obtain a classification result; compare the classification result with the classification label corresponding to the pre-set test data matrix to obtain the classification accuracy; when the classification accuracy is greater than or equal to the test accuracy threshold, use the trained SVM model as the final SVM model; otherwise, repeat S41 and S42 until the classification accuracy is greater than or equal to the test accuracy threshold.

[0075] Preferably, the step of constructing an electric signal data hierarchy system based on the characteristics of the electric signal and performing frequency modulation management on the classified electric signal data based on the constructed electric signal data hierarchy system comprises the following steps:

[0076] The electrical signal data level system is set based on the amplitude range of the muscle electrical signal, with each 0.1mV being set as one level;

[0077] Set the electrical signal frequency modulation management threshold. When the classified electrical signal data is lower than the set evaluation threshold, increase the sampling frequency of the electrical signal data. When the classified electrical signal data is higher than the set evaluation threshold, reduce the sampling frequency of the electrical signal data.

[0078] This embodiment also discloses a system for adjusting the sampling frequency of electromyographic signals, which is used to implement a method for identifying and evaluating mathematical symbols in ancient books. The system includes: a data acquisition device, a data processing module, a data analysis module, a feature extraction module, a classification module, and a frequency modulation management module;

[0079] The data acquisition device includes a detection electrode device and a differential amplifier, which is used to collect and amplify the electrical signal data generated by the muscle in real time;

[0080] The data processing module is used to process the electrical signal data collected in real time to obtain processed electrical signal data;

[0081] The data analysis module is used to analyze the processed electrical signal data;

[0082] The feature extraction module is used to extract features from the analyzed electrical signal data;

[0083] The classification module is used to construct a classification model based on the electrical signal data after feature extraction, and classify the electrical signal data based on the constructed classification module;

[0084] The frequency modulation management module is used to construct an electrical signal data hierarchy system, classify the classified electrical signal data, and perform frequency modulation management based on the classification results.

[0085] (3) Beneficial effects

[0086] Compared with the prior art, the present invention provides a method for adjusting the sampling frequency of electromyographic signals, which has the following beneficial effects:

[0087] 1. The invention collects the electrical signal data generated by the muscle in real time through the installed electrode detection equipment, and constructs an electrical signal data matrix, and processes the collected electrical signal data matrix through a data processing method; at the same time, the processed electrical signal data matrix is ​​analyzed by a data analysis method; and the analyzed electrical signal data is subjected to feature extraction by a feature extraction algorithm; after the extraction is completed, the electrical signal data after feature extraction is summarized to construct an SVM model, and the constructed SVM model is trained by a training optimization method to obtain a final SVM model, and the electrical signal data after feature extraction is classified based on the final SVM model, and finally an electrical signal data hierarchy system is constructed, and the classified electrical signal data is frequency-modulated based on the constructed electrical signal data hierarchy system, thereby improving the accuracy of electromyographic signal sampling.

[0088] 2. This invention completes the collection of electromyographic signals through two groups of electrodes and a differential amplifier at each collection point in the installed electrode detection device, and at the same time determines the electric signal data matrix by setting the collection time point and constructing the data matrix, thereby quantifying the electric signal data collection.

[0089] 3. This invention removes the noise in the collected electrical signal data matrix by using a wavelet function noise reduction method to obtain a noise-reduced electrical signal data matrix. At the same time, the electrical signal data in the noise-reduced electrical signal data matrix is ​​filtered by using the signal-to-noise ratio as a noise reduction evaluation index standard, thereby improving the reliability of electrical signal data processing.

[0090] 4. The invention analyzes the processed electrical signal data matrix by adopting an amplitude keying modulation method and determines the amplitude range of the electrical signal data in the processed electrical signal data matrix. After determining the amplitude range, the electrical signal data in the processed electrical signal data matrix is ​​extracted by a feature extraction method, thereby improving the accuracy of the electrical signal feature extraction.

[0091] 5. This invention constructs an SVM model by summarizing the electrical signal data after feature extraction, and trains the constructed SVM model through a training optimization method to obtain a final SVM model. The electrical signal data after feature extraction is classified and processed based on the final SVM model, thereby improving the intelligence of the electrical signal data classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 It is a structural diagram of the method for adjusting the myoelectric signal sampling frequency of the present invention. DETAILED DESCRIPTION

[0093] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0094] Example 1

[0095] See also Figure 1 This embodiment discloses a method for adjusting the sampling frequency of an electromyographic signal, which specifically includes the following steps:

[0096] S1. Installing a detection electrode device on the muscle surface and collecting electrical signal data generated by the muscle in real time through the installed electrode detection device; and obtaining a matrix of the collected electrical signal data through a matrix construction method;

[0097] S2. Processing the collected electrical signal data matrix by a data processing method to obtain a processed electrical signal data matrix;

[0098] S21, removing noise from the collected electrical signal data matrix by a data noise reduction method to obtain a noise-reduced electrical signal data matrix;

[0099] S22, filtering the noise-reduced electrical signal data by a data filtering method to obtain a filtered electrical signal data matrix;

[0100] S23, setting the filtered electrical signal data matrix as the processed electrical signal data matrix;

[0101] S3. Analyze the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data; and simultaneously extract features from the analyzed electrical signal data using a feature extraction algorithm, and output the feature-extracted electrical signal data;

[0102] S4, summarizing the electrical signal data after feature extraction to construct an SVM model, and training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the electrical signal data after feature extraction based on the final SVM model to obtain classified electrical signal data;

[0103] S5. Constructing an electrical signal data hierarchy system based on the characteristics of the electrical signal, and performing frequency modulation management on the classified electrical signal data based on the constructed electrical signal data hierarchy system;

[0104] Further, see Figure 1The method includes installing a detection electrode device on the muscle surface and collecting the electrical signal data generated by the muscle in real time through the installed electrode detection device; and obtaining the collected electrical signal data matrix through a matrix construction method, which includes the following steps:

[0105] S11, each collection point in the installed electrode detection device includes two sets of electrodes and a differential amplifier; the electrical signals collected by the two sets of electrodes are set as a set of electrical signal data generated by the muscles in real time; after the electrical signal data collection is completed, the collected electrical signal data is amplified by the differential amplifier;

[0106] Furthermore, the electrode detection device is set to collect a set of time points a={a1, a2, ..., a i ,...,a a′}, a i represents the set i-th acquisition time point, and a′ represents the total number of set acquisition time points;

[0107] Furthermore, the expression of the collected electrical signal is as follows:

[0108] y(t)=x(t)-x(t-τ);

[0109] Where y(t) represents the electrical signal data generated by a group of muscles collected at time t, x(t) represents the electrical signal collected by the electrode at time t, and τ represents the time it takes for the electrical signal to be transmitted from one electrode to another.

[0110] S12. Summarize the electrical signal data collected at each set collection time point to obtain a first data matrix; and use the first data as the electrical signal data matrix, as shown below:

[0111]

[0112] Among them, y(a ia' ) represents the data collected by the i-th collection point in the electrode detection device at the a'th time point;

[0113] Further, see Figure 1 , removing noise from the collected electrical signal data matrix by a data denoising method to obtain the electrical signal data matrix after denoising includes the following steps:

[0114] S211, transforming the electrical signal data through a wavelet transform function;

[0115] The wavelet transform function is as follows:

[0116]

[0117] Where m is the scale factor, n is the displacement factor, and f(t) is the square integrable function. is the mother wavelet, W f (m,n) is the wavelet transform function;

[0118] S212: After the transformation is completed, a threshold is set to perform noise reduction processing on the electrical signal data in the electrical signal data matrix;

[0119] The noise expression of the electrical signal data in the electrical signal data matrix is ​​as follows:

[0120] y(t)=f(t)+σ*e(t);

[0121] Where y(t) is the electrical signal data in the electrical signal data matrix, σ is the standard deviation of the noise, and e(t) is the noise;

[0122] The electric signal data after wavelet transformation is divided by the threshold function, and the electric signal data smaller than the threshold is adjusted to 0;

[0123] Threshold function:

[0124]

[0125] Among them, λ is the set threshold, y(t) j,k is the amplitude of the electrical signal data, j represents the horizontal coordinate of the amplitude, k represents the vertical coordinate of the amplitude, sign is the sign function, is the electrical signal data after noise reduction;

[0126] Further, see Figure 1 Filtering the noise-reduced electrical signal data by a data filtering method to obtain a filtered electrical signal data matrix includes the following steps:

[0127] Setting the signal-to-noise ratio as a noise reduction evaluation index standard, and filtering the noise-reduced electrical signal data based on the set noise reduction evaluation index standard;

[0128] The signal-to-noise ratio formula is as follows:

[0129]

[0130] Where SNR stands for signal-to-noise ratio, Represents the electrical signal data after noise reduction;

[0131] The higher the signal-to-noise ratio is set, the stronger the noise suppression effect is;

[0132] Set a signal-to-noise ratio threshold. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is less than the set signal-to-noise ratio threshold, it indicates that the noise reduction result is unqualified. The electrical signal data with unqualified noise reduction results are deleted and filtered. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is greater than or equal to the set signal-to-noise ratio threshold, it indicates that the noise reduction result is qualified.

[0133] Summarizing the filtered electrical signal data to form a filtered electrical signal data matrix;

[0134] Further, see Figure 1 , analyzing the processed electric signal data matrix by a data analysis method to obtain analyzed electric signal data; at the same time, extracting features from the analyzed electric signal data by a feature extraction algorithm, and outputting the feature-extracted electric signal data includes the following steps:

[0135] S31, analyzing the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data;

[0136] Performing modulation analysis on the processed electric signal data matrix using an amplitude keying modulation method to determine the amplitude range of the electric signal data in the processed electric signal data matrix;

[0137] The time domain expression of amplitude keying modulation is:

[0138]

[0139] Among them, the pulse width of the electrical signal d is R d , S MASK (t) is the time domain of amplitude keying modulation, h η is the value of the ηth electrical signal, f k Expressed as the carrier frequency of the kth electrical signal;

[0140] The time domain of the amplitude keying modulation is summed to obtain the electrical signal data after the amplitude keying modulation analysis;

[0141]

[0142] Further, summarizing the electrical signal data after amplitude keying modulation analysis to determine the amplitude range of the electrical signal data in the processed electrical signal data matrix;

[0143] S32, after determining the amplitude range of the electrical signal data in the processed electrical signal data matrix, simultaneously performing feature extraction on the analyzed electrical signal data using a feature extraction algorithm;

[0144] S321, dividing the electrical signal data in the processed electrical signal data matrix into K data blocks of uniform size, and using the divided data blocks as input to the convolutional neural network;

[0145] The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer;

[0146] S322: Based on the divided data blocks, set the convolution kernel size and move it on the input image data blocks according to the set step size, and perform convolution calculation on the data in the data blocks in the corresponding area during the movement, so as to achieve feature extraction for each data block;

[0147] The convolution calculation formula is as follows:

[0148] F = y × w + b;

[0149] Among them, y represents the input data block, w represents the weight of the corresponding convolution kernel, b represents the bias value, and F represents the output feature;

[0150] Furthermore, it is assumed that the convolution layer and the pooling layer are continuously stacked, and the output of the upper layer is the input of the lower layer. After all the convolution and pooling operations are completed, the output image data block features are passed to the fully connected layer;

[0151] The output image data block features are expanded and combined through the fully connected layer to obtain a feature data and save it;

[0152] Further, see Figure 1 , summarizing the electrical signal data after feature extraction to construct an SVM model, and at the same time training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the electrical signal data after feature extraction based on the final SVM model to obtain the classified electrical signal data, including the following steps:

[0153] S41, constructing an initial SVM model; setting the kernel function used by the initial SVM model to be a Gaussian kernel function; as follows,

[0154] The mathematical expression of the support vector machine algorithm is as follows:

[0155]

[0156] Among them, w represents the weight, b represents the bias value, and φ(c) represents a nonlinear mapping from low-dimensional to high-dimensional space. Represents a sample The classification results;

[0157] in,

[0158] Where, represents the Gaussian kernel function; Represents data in the SVM classification space; represents the core of the Gaussian kernel function; express and The Euclidean distance between them; d represents the range of the Gaussian kernel function;

[0159] S42, set the training data ratio to e; divide the electrical signal data after feature extraction according to the training data ratio e to obtain a training data matrix and a test data matrix; set the maximum number of iterations to e' and the training error threshold to

[0160] The training data matrix is ​​input into the initial SVM model to train the initial SVM model. When the number of training iterations is greater than or equal to e' or the training error is less than When , the training of the initial SVM model is stopped to obtain a trained SVM model;

[0161] S43, setting a test accuracy threshold, inputting the test data matrix into the trained SVM model for classification to obtain a classification result; comparing the classification result with the classification label corresponding to the pre-set test data matrix to obtain a classification accuracy; when the classification accuracy is greater than or equal to the test accuracy threshold, using the trained SVM model as the final SVM model; otherwise, repeating S41 and S42 until the classification accuracy is greater than or equal to the test accuracy threshold;

[0162] Further, see Figure 1 , constructing an electric signal data hierarchy system based on the characteristics of the electric signal, and performing frequency modulation management on the classified electric signal data based on the constructed electric signal data hierarchy system includes the following steps:

[0163] The electrical signal data level system is set based on the amplitude range of the muscle electrical signal, with each 0.1mV being set as one level;

[0164] Set the threshold for the management of the frequency modulation of the electric signal. When the classified electric signal data is lower than the set evaluation threshold, increase the sampling frequency of the electric signal data. When the classified electric signal data is higher than the set evaluation threshold, reduce the sampling frequency of the electric signal data.

[0165] Example 2

[0166] In a specific embodiment, the system for adjusting the sampling frequency of electromyographic signals is used to implement a method for identifying and evaluating mathematical symbols in ancient books. The system includes: a data acquisition device, a data processing module, a data analysis module, a feature extraction module, a classification module, and a frequency modulation management module;

[0167] The data acquisition device includes a detection electrode device and a differential amplifier, which is used to collect and amplify the electrical signal data generated by the muscle in real time;

[0168] The data processing module is used to process the electrical signal data collected in real time to obtain processed electrical signal data;

[0169] The data analysis module is used to analyze the processed electrical signal data;

[0170] The feature extraction module is used to extract features from the analyzed electrical signal data;

[0171] The classification module is used to construct a classification model based on the electrical signal data after feature extraction, and classify the electrical signal data based on the constructed classification module;

[0172] The frequency modulation management module is used to construct an electrical signal data hierarchy system, classify the classified electrical signal data, and perform frequency modulation management based on the classification results.

[0173] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for adjusting the sampling frequency of an electromyographic signal, characterized in that: The following steps are involved: S1. Installing a detection electrode device on the muscle surface and collecting electrical signal data generated by the muscle in real time through the installed electrode detection device; At the same time, the collected electrical signal data matrix is ​​obtained through the matrix construction method; S2. Processing the collected electrical signal data matrix by a data processing method to obtain a processed electrical signal data matrix; S21, removing noise from the collected electrical signal data matrix by a data noise reduction method to obtain a noise-reduced electrical signal data matrix; S22, filtering the noise-reduced electrical signal data by a data filtering method to obtain a filtered electrical signal data matrix; S23, setting the filtered electrical signal data matrix as the processed electrical signal data matrix; S3. Analyze the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data; and simultaneously extract features from the analyzed electrical signal data using a feature extraction algorithm, and output the feature-extracted electrical signal data; S4. Summarizing the electrical signal data after feature extraction to construct an SVM model, training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the electrical signal data after feature extraction based on the final SVM model to obtain classified electrical signal data; S5. Construct an electrical signal data hierarchy system based on the characteristics of the electrical signal, and perform frequency modulation management on the classified electrical signal data based on the constructed electrical signal data hierarchy system.

2. A method for adjusting the sampling frequency of an electromyographic signal according to claim 1, characterized in that: The detection electrode device is installed on the muscle surface, and the electrical signal data generated by the muscle is collected in real time through the installed electrode detection device; At the same time, obtaining the collected electrical signal data matrix by the matrix construction method includes the following steps: S11, each collection point in the installed electrode detection device includes two sets of electrodes and a differential amplifier; the electrical signals collected by the two sets of electrodes are set as a set of real-time collection of electrical signal data generated by the muscle; after the electrical signal data collection is completed, the collected electrical signal data is amplified by the differential amplifier; Set the electrode detection equipment collection time point set a={a1,a2,...,a i ,...,a a′ }, a i represents the set i-th acquisition time point, and a′ represents the total number of set acquisition time points; S12. Summarize the electrical signal data collected at each set collection time point to obtain a first data matrix; and use the first data as the electrical signal data matrix.

3. A method for adjusting the sampling frequency of an electromyographic signal according to claim 1, characterized in that: The method of removing noise from the collected electrical signal data matrix by using a data noise reduction method to obtain a noise-reduced electrical signal data matrix comprises the following steps: S211, transforming the electrical signal data through a wavelet transform function; The wavelet transform function is as follows: Where m is the scale factor, n is the displacement factor, and f(t) is the square integrable function. is the mother wavelet, W f (m,n) is the wavelet transform function; S212: After the transformation is completed, a threshold is set to perform noise reduction processing on the electrical signal data in the electrical signal data matrix; The noise expression of the electrical signal data in the electrical signal data matrix is ​​as follows: y(t)=f(t)+σ*e(t); Where y(t) is the electrical signal data in the electrical signal data matrix, σ is the standard deviation of the noise, and e(t) is the noise; The electric signal data after wavelet transformation is divided by the threshold function, and the electric signal data smaller than the threshold is adjusted to 0.

4. The method for adjusting the sampling frequency of an electromyographic signal according to claim 1, wherein: The filtering of the noise-reduced electrical signal data by a data filtering method to obtain a filtered electrical signal data matrix comprises the following steps: Setting the signal-to-noise ratio as a noise reduction evaluation index standard, and filtering the noise-reduced electrical signal data based on the set noise reduction evaluation index standard; The signal-to-noise ratio formula is as follows: Where SNR stands for signal-to-noise ratio, Represents the electrical signal data after noise reduction; The higher the signal-to-noise ratio is set, the stronger the noise suppression effect is; Set a signal-to-noise ratio threshold. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is less than the set signal-to-noise ratio threshold, it indicates that the noise reduction result is unqualified. The electrical signal data with unqualified noise reduction results are deleted and filtered. When the signal-to-noise ratio calculated from the noise-reduced electrical signal data is greater than or equal to the set signal-to-noise ratio threshold, it indicates that the noise reduction result is qualified. The filtered electrical signal data are aggregated to form a filtered electrical signal data matrix.

5. The method for adjusting the sampling frequency of an electromyographic signal according to claim 1, wherein: The method of analyzing the processed electric signal data matrix by a data analysis method to obtain analyzed electric signal data; and extracting features from the analyzed electric signal data by a feature extraction algorithm, and outputting the feature-extracted electric signal data comprises the following steps: S31, analyzing the processed electrical signal data matrix using a data analysis method to obtain analyzed electrical signal data; S32. After determining the amplitude range of the electrical signal data in the processed electrical signal data matrix, feature extraction is performed on the analyzed electrical signal data using a feature extraction algorithm.

6. The method for adjusting the sampling frequency of an electromyographic signal according to claim 5, wherein: Analyzing the processed electrical signal data matrix by a data analysis method to obtain analyzed electrical signal data comprises the following steps: Performing modulation analysis on the processed electric signal data matrix using an amplitude keying modulation method to determine the amplitude range of the electric signal data in the processed electric signal data matrix; The time domain expression of amplitude keying modulation is: Among them, the pulse width of the electrical signal d is R d , S MASK (t) is the time domain of amplitude keying modulation, h η is the value of the ηth electrical signal, f k Expressed as the carrier frequency of the kth electrical signal; The time domain of the amplitude keying modulation is summed to obtain the electrical signal data after the amplitude keying modulation analysis; The electric signal data after amplitude keying modulation analysis is summarized to determine the amplitude range of the electric signal data in the processed electric signal data matrix.

7. The method for adjusting the sampling frequency of an electromyographic signal according to claim 5, wherein: described S321, dividing the electrical signal data in the processed electrical signal data matrix into K data blocks of uniform size, and using the divided data blocks as input to the convolutional neural network; The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer; S322: Based on the divided data blocks, set the convolution kernel size and move it on the input image data blocks according to the set step size, and perform convolution calculation on the data in the data blocks in the corresponding area during the movement, so as to achieve feature extraction for each data block; The convolution calculation formula is as follows: F = y × w + b; Among them, y represents the input data block, w represents the weight of the corresponding convolution kernel, b represents the bias value, and F represents the output feature.

8. The method for adjusting the sampling frequency of an electromyographic signal according to claim 1, wherein: The method of aggregating the electrical signal data after feature extraction to construct an SVM model, training the constructed SVM model through a training optimization method to obtain a final SVM model, and classifying the electrical signal data after feature extraction based on the final SVM model to obtain classified electrical signal data includes the following steps: S41, constructing an initial SVM model; setting the kernel function used by the initial SVM model to be a Gaussian kernel function; The mathematical expression of the support vector machine algorithm is as follows: Among them, w represents the weight, b represents the bias value, and φ(c) represents a nonlinear mapping from low-dimensional to high-dimensional space. Represents a sample The classification results; S42, set the training data ratio to e; divide the electrical signal data after feature extraction according to the training data ratio e to obtain a training data matrix and a test data matrix; set the maximum number of iterations to e' and the training error threshold to The training data matrix is ​​input into the initial SVM model to train the initial SVM model. When the number of training iterations is greater than or equal to e' or the training error is less than When , the training of the initial SVM model is stopped to obtain a trained SVM model; S43. Set a test accuracy threshold, input the test data matrix into the trained SVM model for classification, and obtain a classification result; compare the classification result with the classification label corresponding to the pre-set test data matrix to obtain the classification accuracy; when the classification accuracy is greater than or equal to the test accuracy threshold, use the trained SVM model as the final SVM model; otherwise, repeat S41 and S42 until the classification accuracy is greater than or equal to the test accuracy threshold.

9. The method for adjusting the sampling frequency of an electromyographic signal according to claim 1, wherein: The step of constructing an electric signal data hierarchy system based on the characteristics of the electric signal and performing frequency modulation management on the classified electric signal data based on the constructed electric signal data hierarchy system includes the following steps: The electrical signal data level system is set based on the amplitude range of the muscle electrical signal, with each 0.1mV being set as one level; Set the electrical signal frequency modulation management threshold. When the classified electrical signal data is lower than the set evaluation threshold, increase the sampling frequency of the electrical signal data. When the classified electrical signal data is higher than the set evaluation threshold, reduce the sampling frequency of the electrical signal data.

10. A method for adjusting the sampling frequency of an electromyographic signal according to any one of claims 1 to 9, characterized in that: Also includes: Data acquisition equipment, data processing module, data analysis module, feature extraction module, classification module and frequency modulation management module; The data acquisition device includes a detection electrode device and a differential amplifier, which is used to collect and amplify the electrical signal data generated by the muscle in real time; The data processing module is used to process the electrical signal data collected in real time to obtain processed electrical signal data; The data analysis module is used to analyze the processed electrical signal data; The feature extraction module is used to extract features from the analyzed electrical signal data; The classification module is used to construct a classification model based on the electrical signal data after feature extraction, and classify the electrical signal data based on the constructed classification module; The frequency modulation management module is used to construct an electrical signal data hierarchy system, classify the classified electrical signal data, and perform frequency modulation management based on the classification results.

Images