New Gabor-DCT Feature Extraction Technique for Facial Expression Recognition

Facial Expression Recognition have vital research applications for human machine interaction and wide research issues are need to be resolved in this application of pattern recognition. Feature Extraction is a key stage of Facial Expression recognition on which accuracy of system depends so research issues is to increase accuracy by optimizing the feature extraction stage of expression Recognition System. In the Gabor filter feature selection technique, Gabor equation is projected on facial image using different angles but it generates a high dimension Gabor coefficient matrix of redundant features. The redundancy of features is responsible for increasing the confusion and reducing the accuracy of facial expression recognition system. For increasing the accuracy, redundancy of features should be reduced using filtering process. The proposed Gabor DCT filters technique for Facial Expression recognition system reduces redundant features of Gabor matrices using average DCT filtering technique effectively and Gabor features are optimized towards enhancing the accuracy for facial expression recognition.