文档介绍：
Human Gait Identification Based on Difference Image#
Chen Jinyan*
(School puter Software Tianjin University)
5
10
15
20
Abstract: Human gait recognition aims to identify people by their walking style. In this paper a
difference image based human gait identification method is proposed. For each human gait images
sequence, gauss model based background estimation is used to segment frames of the sequence to
obtain the silhouette images with less noise. paring the difference of two adjacent silhouettes in
the images sequence, we can get a difference images sequence. Every difference image in the
difference sequence indicates the body moving feature during ones walking. By projecting every
difference image to Y axis or X axis we can get two feature vectors. Project every difference images of
the whole walking images in one walking cycle we can get two matrixes. These two matrixes indicate
the style of ones walking. Then Two-Dimensional ponent Analysis(2DPCA) is used to
transform the above matrixes to two vectors while at the same time keep the maximum separability.
Finally the similarity of two human gait images is calculated by the Euclidean distance of the two
vectors. Experimental results on a gait database of 124 people show that the rank 5 identification
accuracy can achieve 92%.
Key words: image processing; human gait; gait identification; ponent analysis;
difference image; visual surveillance
0 Introduction

About Gait Based Human Identification
Human gait is the manner of one’s walking, which was firstly studied in medical field for human
healthiness estimation [1-3]. Doctors analyzed the human gait to find out whether their patients have
health problem. Then researchers [4] found that the age and gender will influence gait. Later
25
researchers [5-7]
found that just like fingerprint, iris and face features, almost everyone has his own
walking features. So it was believed that gait could also be used as a biological feat