Open AccessDetection of human emotions using features based on discrete wavelet transforms of EEG signals
Author(s) -
The Jaswini S,
K. M. Ravikumar
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i1.9.9746
Subject(s) - support vector machine , pattern recognition (psychology) , artificial intelligence , computer science , electroencephalography , artificial neural network , emotion classification , affective computing , wavelet , discrete wavelet transform , entropy (arrow of time) , random forest , speech recognition , feature vector , k nearest neighbors algorithm , wavelet transform , psychology , physics , quantum mechanics , psychiatry
Affective computing is an emerging area of research in human computer interaction where researchers have developed automated assessment of human emotion states using physiological signals to establish affective human compute interactions. In this paper wepresent an efficient algorithm for emotion recognition using EEG signals for the data acquired by audio- video stimuli. The desired frequency bands are extracted using discrete wavelet transforms. The Statistical features, Hjorth parameters, differential entropy and wavelet features are extracted. Artificial neural networks, Support Vector Machine (SVM) and K- nearest neighbor are used on the extracted feature set to develop prediction models and to classify intofour emotion states likeclam, happy, fear and sad .These Artificial neural network models are evaluated on the acquired dataset and emotions are classified into four different states with over all accuracy of 86.36%.The classification rate of calm, happy, fear and sad states are 90.9%, 63.63%, 90.90 and 100 % respectively.