基于多类特征融合的步态识别算法  

A Feature Fusion based Gait Recognition Algorithm

作　　者：纪阳阳[1] 赵辉[1] 张晓利[1] 

机构地区：[1]山东大学信息科学与工程学院,山东济南250100

出　　处：《电气电子教学学报》2009年第5期67-70,共4页Journal of Electrical and Electronic Education

摘　　要：步态识别作为一种新兴的生物认证技术,是指通过人的走路姿势来识别人的身份。由于步态受到很多因素的制约,基于单个特征识别率很低,而且不同的特征其特征类型和度量尺度不同。本文提出一种将人体轮廓特征、肢体角度特征、反射对称特征相融合的方法,得到一个联合特征矢量,并采用最近邻模糊分类器进行识别。实验结果表明,本文算法可以解决不同类别的特征融合问题,具有较好的识别性能。As a new technology of recognition based on biologic character, gait recognition is to discriminate individuals by analysis of gait pattern. Because of the impact of external factors, the recognition rate based on a single feature is unsatisfactory. However, different features have different data types and scales. An efficient gait recognition algorithm based on the fusion of body contour, limbs angles and reflective symmetry feature is presented in this paper. Different features have different weights. For this method, a nearest neighbor fuzzy classifier is used to classify subjects. Experimental results show that the proposed algorithm has effective recognition performance.

关 键 词：步态识别 傅立叶描述子 反射对称因子 特征融合 最近邻模糊分类器 

分 类 号：TB115[理学—数学]