Comparison of different wavelet features from EEG signals for classifying human emotions

In recent years, estimation of human emotions from Electroencephalogram (EEG) signals plays a vital role on developing intellectual Brain Computer Interface (BCI) devices. In this work, we have collected the EEG signals using 64 channels from 20 subjects in the age group of 21
_~
39 years for determining discrete emotions (happy, surprise, fear, disgust, and neutral) under audio-visual induction (video/film clips) stimuli. Surface Laplacian filtering is used to preprocess the EEG signals and decomposed into five different EEG frequency bands (delta, theta, alpha, beta, and gamma) using Wavelet Transform (WT). The statistical features are derived from all these five frequency bands are considered for classifying the emotions using two linear classifiers (K Nearest Neighbor (KNN) & Linear Discriminant Analysis (LDA)). The main objective of this work is to consider a selected number of 24 channels for assessing emotions from the original EEG channels. There are three different wavelet functions (¿db8¿, ¿sym8¿, and ¿coif5¿) are used to derive the linear and non linear features for emotion classification. The validation of statistical features is performed using 5 fold cross validation. In this work, KNN outperforms LDA by offering a maximum average classification rate of 79.174 %. Finally we present the average and individual classification rate of emotions over various statistical features on three different wavelet functions for justifying the performance of our emotion recognition system.