An Improved Average Gabor Wavelet Filter Feature Extraction Technique for Facial Expression Recognition

Facial Expression Recognition has been a very important topic for research in computer pattern recognition and currently there is no method of facial Expression recognition system that have 100% recognition rate. So research issues are to improve recognition rate by improving the pre-processing of datasets, improving the feature extraction method and using the best classifier for face recognition.The purpose of this research work is to increase the recognition rate for face expression recognition system by improving feature extraction method. Feature extraction is the key step on which recognition rate depends for facial gesture recognition. For increasing the recognition rate features using different ways or projection should be extract but there is probability of increasing of redundancy which can be responsible of reducing the recognition rate. High dimension and high redundancy is a problem issue for Gabor while it has maximum variance of features. Dimension and redundancy should be reduced using some technique. The dimension reduction technique for gabor is called filtering so this whole technique is called gabor filter. These filtering technique are sampling, average filtering etc. In the proposed gabor feature extraction technique the gabor features are filtered using wavelet transformation and obtained optimum features from facial dataset which give higher recognition rate compared to average gabor feature extraction technique and gabor sampling filtering feature extraction technique for facial expression