基于运动过程的视觉感知和脑机接口-生物医学工程专业论文.docxVIP

基于运动过程的视觉感知和脑机接口 THE VISUAL PERCEPTION AND THE BRAIN-COMPUTER INTERFACE RESEARCH BASED ON MOTION ABSTRACT Brain computer interface (BCI) is a communication system that does not depend on the brain’s normal output pathways of peripheral nerves and muscles. This technology has become the hot issue of the biomedical engineering research field in nowadays, because it has great value in theory research and practical application. In this study, the experiments in different visual motion stimulations were designed. Classification based on power spectrum and complicity of EEG was made by using the EEG data gained from experiments. The main points of this paper are: Experiments to record EEG after four different visual motion stimulations were designed. After analyzing the principle and the characters of EEG signals, the experiment schemes of four different visual motion stimulations were designed. The EEG datas were recorded at the same time when above stimulations were processed, and then were used in the following off-line analysis. Power spectrum and non-linear complexity of EEG was extracted in research. Because EEG signals are with the properties of the non-stationary and non-linear, the power spectrum AR (Auto-Regressive) and Sample Entropy (SampEn) methods were adopted to extract the features of EEG after four different visual motion stimulations which were used in classification. The results of the SampEn analysis show that the main activities of visual motion perception were appeared in the frontal cortex, occipital and temporal areas. Support Vector Machine (SVM) was used as the classifier. As the realization method of statistical learning theory, SVM not only has simple structure, but also has many improved properties especially extended ability. So it’s a feasible choice to apply SVM in the field of mental tasks classification. In this study, support vector machine was used as classifier. Compared with the classification accuracy rate based on AR model coefficie