Myoelectric gesture recognition method based on RNN-CNN architecture

Abstract

The invention relates to a myoelectric gesture recognition method based on an RNN-CNN architecture. The method comprises the following steps of performing feature extraction on each channel signal byusing an RNN architecture according to a time sequence characteristic of a myoelectric signal, and further extracting a fused feature map by using a CNN architecture, and mainly comprises the following steps of preprocessing the data, using an RNN module to perform preliminary feature extraction on the preprocessed data, using a fusion module to perform fusion processing on an output result of theRNN; using a CNN module to perform feature extraction and analysis on an output result of the fusion module; and using a classification module to judge the input gesture signal by the model output, namely judging which gesture type the electromyographic signal belongs to according to the currently input electromyographic signal. According to the method, the time sequence relevance and characteristics of the data can be effectively extracted, and meanwhile, the gesture recognition rate is improved; an extreme value point selection and splicing method is introduced at a data preprocessing stage, so that the model training time is reduced, and the mutual interference between the channels is avoided; finally, at the fusion stage, the relevance of the multiple channels is utilized, so that theidentification of the electromyographic signals is facilitated.

Description

technical field [0001] The present invention relates to the field of physiological signal recognition, in particular to a myoelectric gesture recognition method based on RNN-CNN architecture. Background technique [0002] EMG signal is a common physiological signal, which is generated by the potential change of muscle fibers, which can reflect muscle movement and provide information on body activity. The myoelectric signals of different gestures usually have certain differences, so the myoelectric signals can be used for the recognition of various hand movements, assisting the research of remote control and robotic arms. In gesture recognition based on electromyographic signals, non-invasive sensors are generally used to obtain electromyographic signals. At present, gesture recognition methods for electromyographic signals can be classified into two categories: gesture recognition methods based on machine learning, and depth-based A learned approach to gesture recognition. ...

AI Technical Summary

Problems solved by technology

[0004] At present, these two types of research still have certain limitations: the recognition method based on machine learning can recognize a small number of gestures, while the recognition method based on deep learning does not simplify the EMG signal on the one hand, which easily leads to a long model training time; In this regard, the timing of EMG signals is not fully considered, and it is easy to affect the further improvement of model accuracy.

Images