Human walking: tracking and analysis

The study of human walking has been of interest to researchers in different disciplines, e.g. modelling of human body, perceiving actions and analysing gait. Our research addresses the question of the existence of diagnostic signatures far human walking from kinematic gait data. This is motivated by the capabilities of humans to perceive gait from reduced spatio-temporal features and characteristics and by early studies of gait analysis aiming to establish ranges of normal values for different gait parameters for men and women from kinematic data. The data for this objective has been collected using 3D motion capture equipment that uses IR-reflective markers placed on the points of interest on the subject's body. We are interested in the motion of the lower extremities of the subjects. The data includes eighty-nine normal and thirty-two pathological subjects. The work presented in this paper comprises tracking of the markers, analysis of the motion, trajectories, extraction of salient features and recognition of gait signatures using a vector quantiser and a clustering mechanism of different groups of subjects-normal and pathological-through interpreting the knowledge obtained from the clusters.