Mental Tasks Classification Of Hands Imaginery Movement

The "Brain Computer Interface (BCI)"system is aimed at developing a technical system with extra communication and control channels that do not depend on the brain's normal output pathways of peripheral nerves and muscles. During these years, the BCI Projects that based on the electroencephalogram (BCI) signals have been the key topics. The purpose of the work in this thesis is to investigate BCI system based on spontaneous electroencephalographic signals, and several distinguishable mental tasks can be interpreted as commands for communication between man and his surroundings. The kind of BCI system would be a promising one because of the simple structure, easy and non-injury EEG signals, no stimulation equipment, short training processing, etc. This paper focuses on some topics described as following.(1) Pre-processing of EEG signalsEEG signals are very weak and are always influenced by eye movements, blinks and muscle, heart and power line noise. And so it is necessary to remove the artifacts and noises in EEG In this paper, Applying Wavelet Transformation and Kalman Filter remove the artifacts, and the excellent result has been achieved.(2) Feature extraction of the EEG signalsIn a BCI system, it is very important to find a meaningful EEG signal representation that contains the remarkable information of different mental actions. Analyzing the origin, the mechanism and the framework of the EEG, Using the wavelet entropy by extraction of wavelet analysis, the approximate entropy, sample entropy and complexity as the features of the EEG signals recorded during imagination of hand movements. Moreover, we compare the performance of these features.(3) Design of the Mental Tasks ClassifierIt is necessary for a BCI system to design a classifier with an excellent performance. For this purpose, different approaches have been investigated in detail. Recent advances in machine learning research have pointed out the advantages of support vector machine (SVM) over other classification techniques. Solid theoretical foundations, good generalization capabilities and easy parameters updating are among the most appealing qualities of SVM for BCI applications. In the experiment, classifiers adopt Fisher discrimination,RBF neural network,K-RBF neural network and support vector machine (SVM). SVM classifier achieves the highest accuracy comparing the others in the classification results.In short, the experimental results show that while the usable information reflecting different conditions of the mental tasks has been properly extracted and the proper classifier has been applied, there is highly accurate discrimination by feature extraction in different mental tasks. Especially the experimental results with Wavelet Entropy and complexity and support vector machine are better. From the experiment results, the wavelet entropy and complexity would be promising methods to extract feature for BCI designs. Meanwhile this BCI design method can be used for classification of several mental tasks. This designed BCI system that can classify different kinds of mental tasks would have a good wide application.