| A brain–computer interface(BCI) is a communication system that does not require any peripheral muscular activity. Indeed, BCI systems enable a subject to send commands to an electronic device only by means of brain activity. Such interfaces can be considered as being the only way of communication for people affected by a number of motor disabilities. The electroencephalogram(EEG) based method is the main interest of current BCI research due to its advantages of low cost, convenient operation and non-invasiveness. Design and implementation of an online system plays an important role in BCI research with the purpose of producing practical devices for real-life clinical application. Compared with offline data analysis, an online BCI has great difficulties in fulfilling real-time processing.From MRI scanner,the posterior hippocampus and posterior parahippocampal gyrus showed greater activity during scenes pattern learning, but not faces pattern learning, whereas perirhinal cortex, anterior hippocampus, and posterior fusiform gyrus were recruited during discrimination learning for faces, but not scenes pattern learning.In this paper, we used two kinds of EEG feature extraction: adaboost strong classifier which constitute by polynomial kernel function of vote perceptron as weak classifiers and penalized logistic regression. The experiments were performed using Emotiv EPOC. The Emotiv headset becomes widely used in consumer BCI application allowing for conducting large-scale EEG experiments in the future.The off-line experiment has shown that the reaction of brain signals were different when subjects stared at faces and scenes pictures, and the recognition rate of two classification algorithms were all above 70%. The on-line control experiment is using the classified EEG signal to control voltage of power over ethernet(POE), which can achieve the stair-like control of LED. With using the intelligent control system of portable dry electrodes EEG cap,the BCI can be used outside the lab, and be into patients’ daily life. |