[1]林海波,王浩,张毅.改进高斯核函数的人体姿态分析与识别[J].智能系统学报,2015,10(03):436-441.[doi:10.3969/j.issn.1673-4785.201405049]
 LIN Haibo,WANG Hao,ZHANG Yi.Human postures recognition based on the improved Gauss kernel function[J].CAAI Transactions on Intelligent Systems,2015,10(03):436-441.[doi:10.3969/j.issn.1673-4785.201405049]
点击复制

改进高斯核函数的人体姿态分析与识别(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第10卷
期数:
2015年03期
页码:
436-441
栏目:
出版日期:
2015-06-25

文章信息/Info

Title:
Human postures recognition based on the improved Gauss kernel function
作者:
林海波 王浩 张毅
重庆邮电大学 智能系统及机器人研究所, 重庆 400065
Author(s):
LIN Haibo WANG Hao ZHANG Yi
Research Center of Intelligent System and Robot, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
关键词:
人体动作姿态识别高斯核函数Kinect欧氏距离测地线距离支持向量机
Keywords:
human posturesrecognitionGauss kernel functionKinectEuclidean distancegeodesic distancesupport vector machines (SVM)
分类号:
TP391.9
DOI:
10.3969/j.issn.1673-4785.201405049
文献标志码:
A
摘要:
为了提高人体动作姿态的识别率,利用Kinect平台构建人体骨骼模型,提出一种基于关节角度的人体姿态表示方法.同时针对传统的高斯核函数中采用欧氏距离计算方法难以完全反映人体关节运动数据样本点与测试点之间位置关系的问题,提出了改进的高斯核函数多类支持向量机(MSVM)人体动作姿态识别方法.在高斯径向基核函数中使用测地线距离代替欧氏距离,建立了基于测地线距离的姿态核函数,采用二叉树方法构建多类支持向量机完成12种上肢姿态的分类.实验结果表明,该算法取得了较好的识别效果,能更加有效识别人体姿态.
Abstract:
In this paper, a method based on the joint angles of human postures is proposed in order to improve the human posture recognition rate through building a human skeleton model on the Kinect platform. For the traditional method of human postures recognition, Euclidean distance is used in Gaussian kernel function, but the positional relationship of sample point and test point of human body joint can not be reflected completely. So the method of improved Gaussian kernel function and multi-class support vector machines (MSVM) is proposed. Using the geodesic distance instead of the Euclidean distance in the Gaussian radial basis kernel function, a posture kernel function based on the geodesic distance is established. Using the binary tree method, a multi-class support vector machine is built to complete classification of 12 kinds of upper limb postures. Experimental results showed that the improved algorithm can identify body postures more effectively than before, achieving a good recognition effect.

参考文献/References:

[1] WEINLAND D, RONFARD R, BOYER E. Free viewpoint action recognition using motion history volumes[J]. Computer Vision and Image Understanding, 2006, 104(2/3): 249-257.
[2] WANG Liang, SUTER D. Informative shape representations for human action recognition[C]//18th International Conference on Pattern Recognition. Hong Kong, China, 2006, 2: 1266-1269.
[3] GIRSHICK R, SHOTTON J, KOHLI P, et al. Efficient regression of general-activity human poses from depth images[C]//2011 IEEE International Conference on Computer Vision (ICCV). Barcelona, Spain, 2011: 415-422.
[4] SEMPENA S, MAULIDEVI N U, ARYAN P R. Human action recognition using dynamic time warping[C]//2011 International Conference on Electrical Engineering and Informatics (ICEEI). Bandung, Indonesia, 2011: 1-5.
[5] VEERARAGHAVAN A, ROY-CHOWDHURY A K, CHELLAPPA R. Matching shape sequences in video with applications in human movement analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(12): 1896-1909.
[6] LUO Ying, WU T D, HWANG J N. Object-based analysis and interpretation of human motion in sports video sequences by dynamic Bayesian networks[J]. Computer Vision and Image Understanding, 2003, 92(2/3): 196-216.
[7] BUCCOLIERI F, DISTANTE C, LEONE A. Human posture recognition using active contours and radial basis function neural network[C]//IEEE Conference on Advanced Video and Signal Based Surveillance. Como, Italy, 2005: 213-218.
[8] BACKES M, HRITCU C, MAFFEI M. Type-checking zero-knowledge[C]//Proceedings of the 15th ACM Conference on Computer and Communications Security. New York, USA: ACM, 2008: 357-370.
[9] 杨庆, 陈桂明, 刘鲭洁, 等. 基于测地线距离的核主元分析在齿轮泵故障识别中应用[J]. 上海交通大学学报, 2011, 45(11): 1632-1636.YANG Qing, CHEN Guiming, LIU Qingjie, et al. Application of kernel principal component analysis based on geodesic distance in pattern recognition of gear pump[J]. Journal of Shanghai Jiao Tong University, 2011, 45(11): 1632-1636.
[10] YUAN Lu, SUN Jian, QUAN Long, et al. Image deblurring with blurred/noisy image pairs[J]. ACM Transactions on Graphics (TOG), 2007, 26(3): 1.
[11] CRISTIANINI N, SHAWE-TAYLOR J. 支持向量机导论[M]. 李国正, 王猛, 曾华军, 译. 北京: 电子工业出版社, 2004.
[12] DU Chengjin, SUN Dawen. Multi-classification of pizza using computer vision and support vector machine[J]. Journal of Food Engineering, 2008, 86(2): 234-242.

相似文献/References:

[1]常征,孟军,施云生,等.多特征融合的lncRNA识别与其功能预测[J].智能系统学报,2018,13(06):928.[doi:10.11992/tis.201806008]
 CHANG Zheng,MENG Jun,SHI Yunsheng,et al.LncRNA recognition by fusing multiple features and its function prediction[J].CAAI Transactions on Intelligent Systems,2018,13(03):928.[doi:10.11992/tis.201806008]

备注/Memo

备注/Memo:
收稿日期:2014-5-22;改回日期:。
基金项目:科技部国际合作项目(2010DFA12160);重庆市工业振兴专项资金资助项目(渝财金[2013]442号).
作者简介:林海波,男,1965年生,副教授,主要研究方向为机器人技术及应用、自动控制技术、模式识别.主持完成省部级及其他科研项目4项,申请国家发明专利2项,重庆市科技创新创业人才支持计划人选,发表学术论文10余篇.王浩,男,1990年生,硕士研究生,主要研究方向为智能系统及机器人、模式识别.张毅,男,1966年生,教授,博士生导师,博士,中国计量测试学会高级会员,重庆市人工智能学会理事,重庆市“322”人才工程第二层次人才,英国Essex大学机器人研究中心访问学者.主要研究方向为机器人技术及应用、生物信号处理及应用、模式识别.主持完成省部级及其他科研项目10余项,申请国家发明专利4项.发表学术论文60余篇,其中被SCI、EI检索30余篇,出版专著1部、教材2部.
通讯作者:王浩. E-mail: haoziwang1990@126.com.
更新日期/Last Update: 2015-07-15