Facial expression recognition using PCA based interface for wheelchair

Human beings transmit a lot of apparent information visually rather than vocally. This consistent expression recognition by machine remains a challenge as of today. To automate recognition of emotional state, machines must be taught to realize facial gestures. The wheelchair is one of the kinds of assistive tools for physically challenged people and elderly people to gain mobility and lead to self-governing life. The use of wheelchair with high maneuverability and navigational intelligence is one of the greatest steps towards the integration of severely disabled people. The aim of this work is to generate a model of a wheelchair command interface that does not require the other´s hands. It includes 3 major modules. They are face detection, facial expression recognition and command generation. The software contains digital image processing for face detection, principal component analysis for facial expression recognition and generating a command signals for interfacing the wheelchair. The algorithm is tested in MATLAB. The stagnant images represent different persons face and facial expressions of a subject, from Indian Face Database and Japanese Female Face Database are used to appraise the effectiveness of the algorithm respectively.