LOCAL BINARY PATTERN FOR EMOTION RECOGNITION USING EEG

Abstract

In the context of Human-Machine Interface and Brain computer interface emotion recognition has demonstrated numerous important roles. However, the majority of currently used emo tion identification techniques perform poorly in recognising emotions, which prevents their widespread use in practical applications. Although emotion recognition based on electroen cephalogram (EEG) has higher priority. Present EEG based emotion recognition system have either lesser performance or higher computation complexity or both. Prevent these problems proposes a Local Binary Pattern (LBP) based emotion recognition system. LBP is a simple yet effective method for texture analysis and classification. In proposed system a LBP of 9 neighbouring pixels are selected, then corresponding binary number will produced by thresholding its value with the value of the centre pixel. This binary code is converted to decimal value. This procedure is iteratively done along the entire signal. The result of this work is that we can efficiently classify emotions with greater classification accuracy and less mathematical complexity than in previous studies.