Low training burden electromyography pattern recognition method and system based on pulse neural network

By employing a low-training-burden electromyography (EMG) pattern recognition method based on spiking neural networks, and utilizing a high-density EMG electrode array and adaptive time-contrast coding, combined with joint decision-making based on voltage, current, and pulse, the problem of heavy training burden and sample dependence in EMG pattern recognition is solved, achieving high-precision gesture recognition.

CN116138794BActive Publication Date: 2026-05-29UNIV OF SCI & TECH OF CHINA
1 Cites 0 Cited by

Patent Information

Application Number
CN202310102352.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-01-19
Publication Date
2026-05-29
Estimated Expiration
2043-01-19

AI Technical Summary

Technical Problem

Existing technologies for electromyography pattern recognition suffer from heavy training burden and sample dependence, making it difficult to achieve high-precision gesture recognition, especially with a small number of samples.

Method used

A low-training-burden electromyographic pattern recognition method based on spiking neural networks is adopted. Signals are collected through a high-density electromyographic electrode array, and an electromyographic pulse sample set is constructed by combining adaptive time-contrast coding and spatial activation information coding. The spiking neural network is trained using a joint decision based on voltage, current and pulse, and an alternative gradient algorithm is used for training.

Benefits of technology

With a small number of samples, the accuracy of gesture recognition was improved, the training burden was alleviated, the overfitting problem was reduced, and high-precision electromyography pattern recognition was achieved.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application relates to a low-training-burden electromyogram pattern recognition method and system based on a pulse neural network, which comprises the following steps: S1, collecting surface electromyogram signals by using a high-density electromyogram electrode array; S2, extracting features after pre-processing the electromyogram signals to obtain electromyogram real-value samples, adding pulses obtained by using adaptive time comparison coding in the time dimension to pulses obtained by using coding based on spatial activation information in the space dimension, and constructing an electromyogram pulse sample set; and S3, inputting the electromyogram pulse sample set into a pulse neural network, jointly deciding based on voltage, current and pulses, training the pulse neural network by using a training algorithm based on alternative gradients, and outputting a predicted gesture category. The method provided by the application can recognize gestures under the constraint of fewer gesture samples, and can effectively relieve the dependence of a classifier on samples.
Need to check novelty before this filing date? Find Prior Art