study of algorithm for gait recognition fased on feature fusion
yang lu-ming,zeng ying，zeng qing-dong,song hong
(college of information science & engineering, central south university, changsha 410083, china)
abstract:this paper proposed a gait recognition algorithm based on the fusion of shape feature and dynamic feature of lower limbs. firstly, used boundary-tracked algorithm to obtain body boundary, then used fourier-based descriptors to describe body contour. secondly, according to the knowledge in body anatomy, computed the coordinates of joints of lower limbs, then got the angles of lower limbs. lastly, with these features fused together to generate representative scores, then used different fusion strategies to compute them. experimental results demonstrate that the performance of proposed algorithm is much better than those based on single feature.......