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.
