<-上一篇/Previous Article 下一篇/Next Article->

[1]王科俊,贲晛烨,刘丽丽.采用Radon变换和二维主成分分析的步态识别算法[J].智能系统学报,2010,5(3):266-271.
　WANG Ke-jun,BEN Xian-ye,LIU Li-li.Gait recognition with Radon transform and D principal component analysis[J].CAAI Transactions on Intelligent Systems,2010,5(3):266-271.

点击复制

采用Radon变换和二维主成分分析的步态识别算法

PDF下载 HTML

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 5 期数: 2010年第3期页码: 266-271 栏目: 学术论文—机器感知与模式识别出版日期: 2010-06-25

Title:: Gait recognition with Radon transform and D principal component analysis

文章编号:: 1673-4785(2010)03-0266-06

作者:: 王科俊¹,贲晛烨^1,2，刘丽丽^1,3; 1.哈尔滨工程大学自动化学院，黑龙江哈尔滨 150001;
2.哈尔滨工业大学交通科学与工程学院，黑龙江哈尔滨 150090;
3.中国科学院沈阳计算技术研究所有限公司，辽宁沈阳 110171

Author(s):: WANG Ke-jun¹, BEN Xian-ye^1,2, LIU Li-li^1,3; 1.College of Automation, Harbin Engineering University, Harbin 150001, China;
2.School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin 150090, China;
3.CAS Shenyang Institute of Computing Technology Co.〖KG-*1/3〗, Ltd, Shenyang 110171, China

关键词:: 步态识别; Radon变换; 二维主成分分析; 模板构造

Keywords:: gait recognition; Radon transform; two dimensional principal component analysis (2DPCA); template construction

分类号:: TP391.41

文献标志码:: A

摘要:: 针对主成分分析算法将图像矩阵转化为向量的维数过高、求取特征向量耗时的问题，综合步态的静态和动态信息，对一个步态周期中的图像进行Radon变换，再通过模板构造，仅用一幅图像来刻画步态特征，接着用二维主成分分析（2DPCA）进行降维.为了验证所提出的算法的有效性，在CASIA步态数据库上进行实验，采用最近邻分类器来测试识别.实验结果表明在特征模板构造时选择合适的频率，采用Radon变换结合列2DPCA进行步态特征提取是有效的.

Abstract:: In the principal component analysis method, concatenating an image matrix often leads to a 1D vector with high dimensionality, which makes it very difficult and timeconsuming to compute the corresponding eigenvectors. By combining static components and dynamic information about the walking style, a novel gait representation was proposed. Gait characteristics were obtained from the Radon transform of gait sequences, where a single image could represent a person’s features by template construction. Then, two dimensional principal component analysis (2DPCA) was used to reduce the dimensions of training and testing data. The nearest neighbor classifier was employed to distinguish the different gaits of human. We tested the proposed gait recognition method on the CASIA gait database. The experimental results demonstrated that, when frequency is chosen properly in template construction, extraction of gait features using the Radon transform and column 2DPCA is very effective. 

参考文献/References:: ［1】王科俊, 侯本博. 步态识别综述［J］. 中国图象图形学报, 2007, 12(7): 11521160.
WANG Kejun， HOU Benbo. A survey of gait recognition［J］. Journal of Image and Graphics, 2007, 12(7): 11521160
［2］WANG Liang, TAN Tieniu. Automatic gait recognition based on statistial shape analysis［J］. IEEE Transcations on Image Processing, 2003, 12(9): 11201131.
［3］URTASUN R, FUA P. 3D tracking for gait characterization and recognition［C］//Proceedings of the 6th International Conference on Automatic Face and Gesture Recognition. Seoul, Korea, 2004: 1722.
［4］黄凤岗, 韩雪花. 基于Radon变换的特征提取在步态识别中的应用［J］. 哈尔滨工程大学学报, 2007, 28(3): 301304.
HUANG Fenggang， HAN Xuehua. Feature extraction based on Radon transform for gait recognition［J］. Journal of Harbin Engineering University, 2007, 28(3): 301304. 
［5］BOULGOURIS N V, CHI Z X. Gait recognition using Radon transform and linear discriminant analysis［J］. IEEE Transation on Image Processing, 2007, 16(3): 731740.
［6］王科俊, 陈薇. 基于Radon变换的步态识别系统［C］//全国模式识别学术会议. 北京, 2007: 223228.
WANG Kejun, CHEN Wei. Feature extraction based on Radon transform for gait recognition［C］//Chinese Conference on Pattern Recognition. Beijing, 2007: 223228. 
［7］YANG Jian, ZHANG DAVIED, FRANGI A F, et al. Twodimensional PCA: a new approach to appearancebased face representation and recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(1): 131137.
［8］王科俊, 贲晛烨, 赵玥. 步态识别中的步态检测与序列预处理［J］. 自动化技术与应用, 2009, 28(8): 6972,79.
?WANG Kejun, BEN Xianye, Zhao Yue. Gait detection and sequence preprocessing for gait recognition［J］. Techniques of Automation and Applications, 2009, 28(8): 6972,79. 
?［9］王科俊, 贲晛烨, 刘丽丽. 基于FanBeam映射的步态识别算法［J］. 哈尔滨工业大学学报, 2008, 40（增刊）:151155.
WANG Kejun, BEN Xianye, LIU Lili. Gait recognition based on fanbeam projection［J］. Journal of Harbin Institute of Technology, 2008, 40(Suppl.): 151155. 
［10］王科俊, 贲晛烨, 孟玮, 等. 基于广义主成分分析的步态识别算法研究［J］. 哈尔滨工程大学学报, 2009, 30(9): 10221028.
?WANG Kejun, BEN Xianye, MENG Wei, et al. Research on a gait recognition algorithm based on generalized principal component analysis［J］. Journal of Harbin Engineering University, 2009, 30(9): 10221028.
［11］YU Shiqi, WANG Liang, HUANG Kaiqi, et al. Gait analysis for human identification in frequency domain［C］//Proceedings of 3rd International Conference on Image and Graphics. Hong Kong, China, 2004: 282285.

相似文献/References:: [1]王彦杰,汪增福.基于光学图像的舰船航迹检测[J].智能系统学报,2007,2(4):46.
　WANG Yan-jie,WANG Zeng-fu.Ship wake detection using optical images[J].CAAI Transactions on Intelligent Systems,2007,2():46.
[2]张元元,吴晓娟,李秀媛,等.平行线约束下的视角无关步态识别算法[J].智能系统学报,2009,4(3):264.
　ZHANG Yuan-yuan,WU Xiao-juan,LI Xiu-yuan,et al.Viewpointindependent gait recognition with parallel line constraints[J].CAAI Transactions on Intelligent Systems,2009,4():264.
[3]贲晛烨,王科俊,刘海洋.核方法的对比研究及在步态识别中的应用[J].智能系统学报,2011,6(1):63.
　BEN Xianye,WANG Kejun,LIU Haiyang.A comparative study on kernel methods and their applications to gait recognition[J].CAAI Transactions on Intelligent Systems,2011,6():63.
[4]高海燕,阮秋琦.正面视角的步态识别[J].智能系统学报,2011,6(2):119.
　GAO Haiyan,RUAN Qiuqi.A gait recognition method based on frontview[J].CAAI Transactions on Intelligent Systems,2011,6():119.
[5]杨静,阮秋琦,李小利.基于频谱分析的Procrustes统计步态识别算法[J].智能系统学报,2011,6(5):432.
　YANG Jing,RUAN Qiuqi,LI Xiaoli,et al.A Procrustes statistical gait recognition algorithm based on spectrum analysis[J].CAAI Transactions on Intelligent Systems,2011,6():432.
[6]贲晛烨,王科俊,马慧.视频下的正面人体身份自动识别[J].智能系统学报,2012,7(1):69.
　BEN Xianye,WANG Kejun,MA Hui.Videobased automatic frontview human identification[J].CAAI Transactions on Intelligent Systems,2012,7():69.
[7]李一波,李昆.双视角下多特征信息融合的步态识别[J].智能系统学报,2013,8(1):74.[doi:10.3969/j.issn.1673-4785.201209033]
　LI Yibo,LI Kun.Gait recognition based on dual view and multiple feature information fusion[J].CAAI Transactions on Intelligent Systems,2013,8():74.[doi:10.3969/j.issn.1673-4785.201209033]
[8]贲晛烨,张鹏,潘婷婷,等.线性插值框架下矩阵步态识别的性能分析[J].智能系统学报,2013,8(5):415.[doi:10.3969/j.issn.1673-4785.201110007]
　BEN Xianye,ZHANG Peng,PAN Tingting,et al.Performance analysis of matrix gait recognition under? linear interpolation framework[J].CAAI Transactions on Intelligent Systems,2013,8():415.[doi:10.3969/j.issn.1673-4785.201110007]
[9]李浩淼,张含笑,邢向磊.联合局部多尺度和全局上下文特征的步态识别[J].智能系统学报,2024,19(4):853.[doi:10.11992/tis.202304004]
　LI Haomiao,ZHANG Hanxiao,XING Xianglei.Gait recognition with united local multiscale and global context features[J].CAAI Transactions on Intelligent Systems,2024,19():853.[doi:10.11992/tis.202304004]

备注/Memo

收稿日期：2009-03-18.
基金项目：国家“863”计划资助项目（2008AA01Z148）.
通信作者：贲晛烨.E-mail:benxianyeye@163.com.
作者简介：
王科俊，男，1962年生，教授、博士生导师、博士，哈尔滨工程大学自动化学院副院长，模式识别与智能系统学科带头人. 主要研究方向为模糊混沌神经网络、自适应逆控制理论、可拓控制、网络智能控制、模式识别、多模态生物特征识别、联脱机指纹考试身份鉴别系统、微小型机器人系统等.完成科研项目20余项，目前在研项目10余项.曾获得部级科技进步二等奖2项，三等奖3项，省高校科学技术一等奖1项、二等奖1项.已授权发明专利1项、公开3项，获国家版权局软件著作权登记1项.发表论文180余篇，出版学术专著3部，国防教材1部，主审教材2部.
贲晛烨，女，1983年生，博士研究生.主要研究方向为模式识别、生物特征识别、智能交通系统.申请专利5项，发表学术论文17篇.
刘丽丽，女，1982年生，硕士，主要研究方向为模式识别、智能控制、工业控制.发表学术论文4篇.

更新日期/Last Update: 2010-07-14

采用Radon变换和二维主成分分析的步态识别算法 PDF下载HTML

备注/Memo

采用Radon变换和二维主成分分析的步态识别算法

PDF下载 HTML