Unlock instant, AI-driven research and patent intelligence for your innovation.

Feature extraction method of motor imagery EEG signal based on weighted compound multi-scale fuzzy entropy

A technology of motor imagery and EEG signals, applied in character and pattern recognition, instruments, calculations, etc., can solve problems that are unreasonable and affect the accuracy of pattern classification, and achieve the effect of improving classification accuracy

Active Publication Date: 2021-06-04
BEIJING UNIV OF TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It is obviously unreasonable to use CMFE to deal with time-varying MI-EEG, which will inevitably affect the accuracy of pattern classification

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feature extraction method of motor imagery EEG signal based on weighted compound multi-scale fuzzy entropy
  • Feature extraction method of motor imagery EEG signal based on weighted compound multi-scale fuzzy entropy
  • Feature extraction method of motor imagery EEG signal based on weighted compound multi-scale fuzzy entropy

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The experimental data used in the present invention comes from the "BCI Competition 2003" competition database. The EEG signals of the three leads C3, Cz, and C4 were recorded, with a sampling frequency of 128 Hz and a band-pass filter of 0.5 to 30 Hz. The acquisition timing diagram is as follows figure 1 As shown, a single experiment lasted 9s, the subjects were in a relaxed state for the first two seconds, a "+" cursor appeared on the screen at t=2s, t=3~9s, and the subjects completed imagining the left or right hand movements according to the prompts . Both the training set and the test set contain 140 experimental data, of which the left and right hand movements are imagined 70 times each.

[0038] like figure 2 As shown, the specific implementation steps of the present invention are:

[0039] (1) Preprocessing of EEG signals

[0040] For MI-EEG signals imagining left and right hand movements, under the category T motor imagery task, suppose for the i-th lea...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a feature extraction method of motor imagery EEG signals based on weighted composite multi-scale fuzzy entropy. First, the composite multi-scale fuzzy entropy time series of MI-EEG is calculated, and the CMFE entropy of different leads of each motor imagination task T is calculated. The optimal time period is determined by the difference and change of the value, and the MI‑EEG signal in this period is further used for feature extraction; then, weight factors are introduced into different sampling points in the CMFE coarse-graining process to obtain τ weighted coarse-grained Then, the fuzzy entropy of each coarse-grained sequence is calculated, and the average value is defined as WCMFE; for various motion imagery tasks T arbitrary lead C i Calculate the WCMFE at a single scale τ; determine the variation range of the scale factor τ, calculate the WCMFE at multiple scales, and sequentially construct the eigenvectors at each scale τ and the eigenvectors F of various motion imaging tasks T T , and further fused into the feature vector F of MI‑EEG, which further improves the classification accuracy.

Description

technical field [0001] The invention belongs to the technical field of EEG signal processing, and is applied to the feature extraction of motor imagery EEG signals in a brain-computer interface (Brain-Computer Interface, BCI) system. WCMFE) method for nonlinear dynamic feature extraction of motor imagery EEG signals. Background technique [0002] When people perform motor imagery, the cerebral cortex will generate motor imagery electroencephalography (MI-EEG) signals with rhythmic activity. Brain Computer Interface (BCI) technology based on motor imagery can be used for training and rehabilitation of patients with motor nerve dysfunction. Since MI-EEG is sensitive to noise and has time-varying and fuzzy characteristics, its feature extraction has become a key issue in BCI-based rehabilitation engineering research. [0003] With the development of nonlinear dynamics, many studies have shown that the brain is a nonlinear dynamic system. Among them, entropy is a common metho...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
CPCG06F2218/08
Inventor 李明爱王若图杨金福
Owner BEIJING UNIV OF TECH