Facial emotion recognition for human-machine interaction using hybrid DWT-SFET feature extraction technique

Facial emotion recognition is the most significant parameter for an efficacious Human Machine Interaction (HMI). It plays a crucial role in interpreting and communicating with the people who have speaking impairments as well as a medium to understand and communicate with infants who cannot emote their feelings verbally. In this paper, we propose a hybrid feature extraction technique consisting of Discrete Wavelet Transform (DWT) accompanied by Shape Feature Extraction Technique (SFET).The features extracted were tested on standard classifiers such as Support Vector Machine (SVM) and K-Nearest Neighbourhood (KNN) classifiers. Facial images from JAFFE and Cohn-Kennedy databases were utilized for training as well as testing. The work shows a very high facial emotion recognition rate of 93.94% and 91.8% with the proposed method for JAFFE and Cohn-Kanade databases respectively.