The feature extraction of motor imagery EEG based on the time-frequency correction

Author

Dongyang, Wang ; Jing, Jin ; Xingyu, Wang

Author_Institution

Key Lab. Adv. Control & Optimization for Chem. Processes of Minist. of Educ., East China Univ. of Sci. & Technol., Shanghai, China

fYear

2012

fDate

25-27 July 2012

Firstpage

3803

Lastpage

3805

Abstract

A novel method based on Time-Frequency correction was presented to extract motor imagery feature from brain signal. The 2b datasets of BCI competition IV were used in this paper. Time-frequency features of right and left hand motor imagery were extracted from train datasets, respectively. The frequency band and time segment were selected based on frequency and time correction. The features were extracted from selected frequency band and time segment. The results showed that the classification accuracy using the features from selected frequency band and time segment is higher than that using the normal band power features.

Keywords

brain-computer interfaces; electroencephalography; feature extraction; medical signal processing; signal classification; time-frequency analysis; brain signal; frequency band; frequency correction; left hand motor imagery; motor imagery EEG; motor imagery feature extraction; normal band power features; right hand motor imagery; time segment; time-frequency correction method; time-frequency features; Brain computer interfaces; Electroencephalography; Electronic mail; Feature extraction; Image segmentation; Optimization; Time frequency analysis; Band Power; Brain-Computer Interface (BCI); Motor Imagery; Time-Frequency Feature;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (CCC), 2012 31st Chinese

Conference_Location

Hefei

ISSN

1934-1768

Print_ISBN

978-1-4673-2581-3

Type

conf

Filename

6390588