Facial Expression Recognition Using Gabor Wavelet & Neural Networks

This paper presents methods for identifying facial expression. The objective of this paper is to present a combination texture oriented method with dimensional reduction for identifying facial expressions. Conventional methods have difficulty in identifying expressions due to change in the shape of the cheek. By using simple two dimensional image analysis , the accuracy of the expression detection becomes difficulty. Without considering the three dimensional analysis, by using texture extraction of the cheek, the researchers are able to increase the accuracy of the expression detection. In order to achieve the expression detection accuracy, Gabor wavelet is used in different angles to extract possible texture of the facial expression. The texture dimension is further reduces by using Fisher's linear discriminant function for increasing the accuracy of the proposed method. Fisher's linear discriminant function from transforming higher dimensional feature vector into two-dimensional vector training and identifying expressions. Different facial expressions considered are angry, disgust, happy, sad, surprise and fear are used. These expressions can be used for security purposes.