Md Alamgir Hossain¹, Debabrata Samanta² and Goutam Sanyal², ¹Calcutta Institute of Technology, India and ²National Institute of Technology, India

Facial expression scrutiny has attracted tremendous consciousness in the area of computer vision because it plays a prime role in the domain of human-machine communication. Smiley face expressions are generated by slimming down of facial muscles, which results in temporally buckled facial features such as eye lids, eye brows, nose, lips, and skin texture. These are evaluated by three characteristics: those portions of the face that will take part for facial action, the intensity of facial actions, and the dynamics of facial actions. In this paper we propose a real-time, accurate, and robust smile detection method based on genetic algorithm. We generated leaf-matrix to extract target expression. Finally, we have compared our methodology with the smile shutter function of Canon Camera. We have achieved better performance than Sony on slight smile.

Leaf-matrix, genetic algorithm, pattern matching, smile detection.