| 用行人轮廓的分布直方图分类和识别步态 |
| |
| 引用本文： | 陈实,马天骏,高有行.用行人轮廓的分布直方图分类和识别步态[J].计算机研究与发展,2009,46(2). |
| |
| 作者姓名： | 陈实 马天骏 高有行 |
| |
| 作者单位： | 1. 浙江万里学院设计学院,浙江宁波,315100;西安电子科技大学计算机学院,西安,710071
2. 西安电子科技大学计算机学院,西安,710071 |
| |
| 基金项目： | 宁波市自然科学基金，陕西省自然科学基金 |
| |
| 摘 要： | 现有的步态识别方法对行人轮廓匹配的鲁棒性差,识别率不高.提出了一种基于轮廓直方图分布的行人步态识别方法.首先提取行人二值轮廓序列;然后通过人体局部轮廓的点分布直方图获取步态周期;继而构造表达帧间关系的周期步态平面,设计一种直方图分布的描述子获得帧姿态特征值,计算出姿态轮廓特征分布间的Jeffery距离,结合动态时间规整技术获取了测试序列和参考序列间的匹配相似度,最终完成了识别.在Soton步态数据库上进行了实验,提出算法的正确识别率可达87.59%,与相关文献的时比分析表明算法是有效的.
|
| 关 键 词： | 步态识别 行为特征 步态分析 行人轮廓 形状匹配 直方图分布 |
Gait Recognition Using Distributions of Silhouette Feature |
| |
| Chen Shi,Ma Tianjun,Gao Youxing.Gait Recognition Using Distributions of Silhouette Feature[J].Journal of Computer Research and Development,2009,46(2). |
| |
| Authors: | Chen Shi Ma Tianjun Gao Youxing |
| |
| Affiliation: | Faculty of Design;Zhejiang Wanli University;Ningbo;Zhejiang 315100;School of Computer Science and Technology;Xidian University;Xi'an 710071 |
| |
| Abstract: | Vision-based human identification at a distance in surveillance has recently gained more attentions. Gait has the advantages of being non-invasive and difficult to conceal,and is also the only perceivable biometric at a distance. This paper introduces a novel feature representation method for gait analysis and recognition applications. The method includes following steps:first,silhouette extraction is performed for each image sequence. Secondly,the distributions of sampled points from the human local silhou... |
| |
| Keywords: | gait recognition behavioral biometrics gait analysis human silhouette shape matching distributions |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
