Classification of EEG Signals for Motor Imagery based on Mutual Information and Adaptive Neuro Fuzzy Inference System

摘要: Electroencephalogram EEG signals based Brain Computer Interface BCI is employed to help disabled people interact better with the environment. are recorded through system translate it control commands. There a large body of literature targeting feature extraction and classification for Motor Imagery tasks. imagery task have several features can be extracted use in classification. However, using more consume running time irrelevant redundant affect performance used classifier. This paper dedicated extracting best vector motor task. work suggests two selection methods on Mutual Information MI including Minimum Redundancy Maximal Relevance MRMR maximal MaxRel. Adaptive Neuro Fuzzy Inference System ANFIS classifier Subtractive clustering method utilized classifications. The suggested applied Competition III dataset IVa IVb II III.