STATISTICAL APPROACH FOR PERSON AUTHENTICATION USING BRAIN SIGNAL
This paper presented the usage of statistics over the set of the features representing the electroencephalogram (EEG) signals. Since classification is more accurate when the pattern is simplified through representation by important features. Feature extraction and selection play an important role in classifying systems such as Deep Maxout networks. Six different techniques such as Empirical Mode Decomposition, Spectral flux, Logarithmic band power, Zero crossing rate, PCA and Differential Entropy are used for extracting features of Brain signal. Then these features are combined into feature vector, which is used as input for the Deep Maxout Network. The aim is to extract appropriate features, which will be helpful for person authentication system.