Simple Approach for Face Expression Recognition Using Dual Classifier for Enhanced Throughput

Abstract

This paper presents a facial expression recognition system using dual classifier approach. The face is
detected using viola jones algorithm. The Gabor filters are used for feature extraction. The dimensionality of the feature
vector is reduced by the Principal Component Analysis (PCA) to remove redundant data that leads to unnecessary
computation cost. The Support Vector Machine (SVM) and Bayesian Classifiers are used sequentially for expression
classification. The performance of the proposed method is tested on public and largely used Cohn -Kanade database.The
experiments show that proposed method gives promising results.