Classifying ECoG-Based motor imagery tasks by feature weighted C-means algorithm based on PSO

摘要：

The electrocorticogram (ECoG) is a feasible source of BCI. Pattern classification for ECoG features is very challenging in case of a classifier that was trained on the first day shall classify data recorded during following days. In this paper, a new method is proposed for classifying ECoG trials during motor imagery. Hilbert-Huang Transform (HHT) marginal spectrum are extracted as classification features. Sliding windows (SW) and a Fisher linear discriminant (FLD) is used for feature dimensionality reduction, C-means and a particle swarm optimization (PSO) algorithm is used to optimize the feature weights. Using 6 channels, the test accuracy of 0.9 is achieved on Data set I of BCI Competition III.

作者: Shengli Xie Jun Lv

作者单位: South China University of Technology Guangzhou, P.R.China 510641

会议类型: 国际会议

会议名称: Fourth International Conference on Impulsive and Hybrid Dynamical Systems(ICIHDS 2007)(第四届国际脉冲和混合动力系统学术会议)

会议地点: 南宁

会议语种:英文

在线出版日期: 2007-07-20（万方平台首次上网日期，不代表论文的发表时间）