A Method of FAM for Pedestrian Behaviour Classification

In this paper, we present the classification of motion, and make a conclusions under specific circumstances. In order to indicate a pedestrian´s movements, a complex number notation based on centroid is proposed. And according to the different sorts of movements, a set of standard image contours are made. Different behaviour matrices based on spatio-temporal are acquired through hidden Markov models (HMM). A procrustes mean shape distance (PMSD) analysis method is presented in order to get the degree to which two contours are resembled. Aa a final result, a learning fuzzy associative memory (FAM) is proposed to infer behavior classification of a walker, and an evaluation of six kinds of gaits involving walking, side walking, jumping, crouching, uphill and downhill is given with a 62.5% recognition rate achieved. Experiment results shows the system computational cost has been effectively reduced ,the robustness of method has been validated by experiments.