Single-time brain wave characteristic extraction and classification method for motion execution

Abstract

The invention discloses a single-time brain wave feature extraction and classification method for motion execution. According to the method, a multi-element empirical mode decomposition algorithm anda common spatial pattern algorithm are combined to process a single-time brain wave in motion execution; and after brain wave signal characteristics representing different motion intentions are extracted, the extracted brain wave signal characteristics are classified, and a motion execution category represented by the single-time motion execution brain wave is obtained. Through the method, the problem that existing multi-classification motion execution tasks are low in classification correct rate can be solved; and through combination of multi-element empirical mode decomposition and a commonspatial pattern, the recognition rate of multi-task brain wave signals is effectively increased, and the complexity in the operation process is lowered.

Description

technical field [0001] The invention belongs to the technical field of EEG signal recognition control, in particular to a method for extracting and classifying single-shot EEG features of exercise execution, a single-shot EEG classification method for exercise execution based on multiple empirical mode decomposition and common spatial patterns, which is used for different A study of pattern recognition in EEG signals for motor execution. Background technique [0002] The brain is the most important physiological organ of the human body. It enables the human body to perform a series of behavioral activities by instructing the neuromuscular. However, many diseases can reduce or even block the transmission function between the brain and nerves, resulting in the loss of muscle activity. Studies have shown that the human brain has plasticity in structure and function, and early and correct rehabilitation training can promote the repair of brain nerves, reduce neuron death, and i...

AI Technical Summary

Problems solved by technology

[0004] In view of the low accuracy rate of the existing multi-type motion execution task classification, the purpose of the present invention is to provide a single-shot EEG feature extraction and classification method for motion execution, which is based on multiple empirical mode decomposition and common spatial patterns. The single-shot EEG signal classification method, the CSP algorithm can extract the characteristics of multi-channel signals from the spatial perspective, and effectively improve the signal-to-noise ratio of the signal; the MEMD method can decompose the nonlinear and non-stationary EEG signals into multiple stable eigenmodes state function, to realize the smooth processing of non-stationary signals, and combine the two methods for feature extraction of EEG signals, which can effectively improve the recognition rate of multi-task EEG signals

Images