AN ANALYSIS OF GREYHOUND GAIT USING WAVELETS

In this paper we present a method of analysing gait of quadrupedal ani mals using wavelets. Time series data such as joint trajectories can b e decomposed by the discrete wavelet transform to represent components of different frequency bandwidth. Differences between two similar tra jectories can be detected by comparing the components of the same band width. We analysed kinematic data of the hindlimbs of three greyhounds with normal gait and with tibial nerve paralysis. Abnormalities in th e gait patterns were detected and quantified by comparing the energy c ontribution of the components that were present at the same level of t he wavelet transform. (C) 1997 Elsevier Science Ltd.