Human Mood Detection Using Genetic Algorithm
International Journal of Modern Trends in Engineering and Research, Aug 27, 2015
Perception modeling of facial expression is the vital step of human-machine interaction, which is... more Perception modeling of facial expression is the vital step of human-machine interaction, which is an important study area of the present time. In this paper I have determined five well known human moods ‘happy’, ‘sad’, ‘angry’, ‘surprise’ and ‘normal’ by extracting three basic facial attributes eye opening (EO), mouth opening (MO) and eyebrow constriction (EBC) from faces of still images. After that using suitable membership function extracted attributes are converted into three fuzzy sets i.e. high, medium and low. Taking facial attributes of well defined moods in initial database automated learning is implemented to construct the knowledgebase of facial expression with the help of genetic algorithm using a suitable objective function. On an average it results 97.6% success in mood detection.
Uploads
Papers by Avishek Gupta