Abstract: Partition based feature extraction is widely used in the pattern recognition and computer vision. This method is robust to some changes like occlusion, background, etc. In this study, partition based technique is used for feature extraction and extension of HMM is used as a classifier. The new introduced multi-stage HMM consists of two layers. In which bottom layer represents the atomic expression made by eyes, nose and lips. Further upper layer represents the combination of these atomic expressions such as smile, fear, etc. Six basic facial expressions are recognised, i.e., anger, disgust, fear, joy, sadness and surprise. Our proposed system is able to get overall accuracy of 82% using JAFFE database.
Mayur Rahul, Narendra Kohli and Rashi Agrawal, 2018. Partition Based Feature Extraction Technique for Facial Expression Recognition Using Multi-Stage Hidden Markov Model. Journal of Engineering and Applied Sciences, 13: 2651-2658.