EEG based brain signal acquisition and analysis II
Lim, Ting Min
Date of Issue2016-05-19
School of Electrical and Electronic Engineering
In recent years, there is an increase of interest in the nature of emotion recognition from EEG (Electroencephalogram) signals. EEG system (consists of g.USBamp, g.SAHARA, g.GAMMAcap) is a very simple and effective tool use for measuring electrical activity of human brain in understanding its complex behavior. The objective of this study is to develop an algorithm using Matlab to automatically classify two different emotional states (Happiness and Sadness) in the EEG signals. To motivate these two emotional states, pictures and background audio were taken from International Affective Picture System and online sources respectively. This experiment was designed with EEG system (a cap with 8 electrodes attached) to record the EEG signals from participants. These signals were then proceed to: i. Pre-processing to get rid of the noise and extract useful signal ii. Feature extraction to extract time-frequency domain features (Discrete Wavelet Transform) and entropy of the EEG signal iii. Feature selection using PCA (Principal Component Analysis) to maximize the performance of a learning algorithm iv. Classification using SVM (Support Vector Machine) to classify the emotional state With this procedures, an average recognition rate up to 71% for two emotional states is achieved.
Final Year Project (FYP)
Nanyang Technological University