A FUZZY APPROACH TO PERSON fIDENTIFICATION USING GAIT

A biometric is a measure taken from a person and used for identity verification or recognition. Gait is a biometric perceptible from a distance even at low resolution. In this paper, we propose a novel method for person identification via gait analysis. A biomechanical model is used for the purpose. The algorithm detects the edge image as a first step. This is followed by the development of line model and extraction of features. As a final step, a fuzzy classifier with reject option is used for the recognition of the person under consideration. In our experiment, we achieved an accuracy of 92%. Thus, our method proves to be very efficient for the purpose.