Recognition method of motor imagery EEG (electroencephalograph) signal based on energy characteristics

Abstract

The invention relates to a recognition method of motor imagery EEG (electroencephalograph) signals based on energy characteristics. The recognition method of the motor imagery EEG signals based on theenergy characteristics includes the steps of collecting EEG signals from subjects as sample sets during the time when the subjects are completely relaxed and when the subjects are conducting motion imagination, denoising the EEG signals of these sample sets and acquiring the average power of the EEG signals, acquiring the energy values of the Mu (8-12 Hz) band and the Beta (18-25 Hz) band of theEEG signals as the eigenvalues during the motion imaging state and the fully relaxed state based on Fourier transform and using the energy spectral density function, and conducting online optimizationof parameters of classification methods to realize the recognition of the motor imagery EEG signals with the characteristics as input and using a support vector machine classification method based ona radial basis kernel function as well as using an improved grid optimization algorithm. The recognition method of the motor imagery EEG signals based on the energy characteristics has the advantagesof consuming short time, guaranteeing high accuracy and providing a new idea for real-time BCI (bulk current injection) implementation.

Description

technical field [0001] The invention relates to the field of electroencephalogram signal processing, in particular to a method for identifying motor imagery electroencephalogram signals based on energy features. Background technique [0002] Electroencephalogram (Electroencephalogram, EEG) is a method that uses electrophysiological indicators to record brain activity. When the brain is active, the synchronous post-synaptic potentials of a large number of neurons are summed and formed. The subject imagines performing an action without producing any physical movement, a process known as motor imagery. The results of functional imaging studies have shown that there is a functional relationship between self-motion and motor imagery, and imaginary movement can activate the corresponding brain regions involved in self-motion. [0003] When a certain area of ​​the cerebral cortex is stimulated by imaginative movement, the metabolism of this area is accelerated, and information pro...

AI Technical Summary

Problems solved by technology

Among them, the experimental method is time-consuming and difficult to obtain the optimal parameters, the grid search method has high precision but is very time-consuming, the bilinear search method is efficient but the accuracy is average, the genetic algorithm and the particle swarm optimization algorithm are more complicated and easy to fall into local optimum

Images