基于Kinect骨骼预定义的体态识别算法

计算机应用 ›› 2014, Vol. 34 ›› Issue (12): 3441-3445.

基于Kinect骨骼预定义的体态识别算法

张丹¹,陈兴文¹,赵姝颖²,李纪伟²,白钰³

1. 大连民族学院创新教育中心,辽宁大连 116600
2. 东北大学信息科学与工程学院,沈阳 110819;
3. 大连民族学院信息与通信工程学院,辽宁大连 116600

收稿日期:2014-06-27 修回日期:2014-09-10 发布日期:2014-12-31 出版日期:2014-12-01
通讯作者: 张丹
作者简介:张丹(1987-),女,辽宁大连人,工程师,硕士,主要研究方向:图像处理、机器视觉;陈兴文(1969-),男,辽宁大连人,教授,主要研究方向:图像处理、模式识别；赵姝颖(1968-),女,辽宁沈阳人,教授,博士,主要研究方向:图像处理、计算机视觉；李纪伟(1987-）,男,河南郑州人,工程师,主要研究方向:计算机视觉、人机交互。
基金资助:
中央高校基础科研基金资助项目;机器人学国家重点实验室开放基金资助项目

Posture recognition method based on Kinect predefined bone

ZHANG Dan¹,CHEN Xingwen¹,ZHAO Shuying²,LI Jiwei²,BAI Yu³

1. Innovation Education Center, Dalian Nationalities University, Dalian Liaoning 116600,China;
2. College of Information Science and Engineering, Northeastern University, Shenyang Liaoning 110819, China;
3. College of Information and Communication Engineering, Dalian Nationalities University, Dalian Liaoning 116600, China

Received:2014-06-27 Revised:2014-09-10 Online:2014-12-31 Published:2014-12-01
Contact: ZHANG Dan

摘要/Abstract

摘要：

针对基于视觉的体态识别对环境要求较高、抗干扰性差等问题,提出了一种基于人体骨骼预定义的识别分类方法。该算法结合Kinect多尺度深度信息和梯度信息检测人体；基于随机森林采用正负样本互限思想识别人体各个部分,根据各部分距离构建人体姿态向量,识别骨架；再根据体态类别,构建最优分类超平面、核函数,采用改进的支持向量机进行体态分类。实验结果表明,所提算法的分类识别准确率可达94.3%,具有实时性好,抗干扰性强,鲁棒性较好等特点。

Abstract:

In view of the problems that posture recognition based on vision requires a lot on environment and has low anti-interference capacity, a posture recognition method based on predefined bone was proposed. The algorithm detected human body by combining Kinect multi-scale depth and gradient information. And it recognized every part of body based on random forest which used positive and negative samples, built the body posture vector. According to the posture category, optimal separating hyperplane and kernel function were built by using improved support vector machine to classify postures. The experimental results show that the recognition rate of this scheme is 94.3%, and it has good real-time performance, strong anti-interference, good robustness, etc.

中图分类号:

TP391

张丹陈兴文赵姝颖李纪伟白钰. 基于Kinect骨骼预定义的体态识别算法[J]. 计算机应用, 2014, 34(12): 3441-3445.

ZHANG Dan CHEN Xingwen ZHAO Shuying LI Jiwei BAI Yu. Posture recognition method based on Kinect predefined bone[J]. Journal of Computer Applications, 2014, 34(12): 3441-3445.

参考文献

［1］RAO C, YILMAZ A, SHAH M. View-invariant representation and recognition of actions［J］. International Journal of Computer Vision, 2002,50(2): 203-226.
［2］PARAMESWARAN V, CHELLAPPA R. View invariants for human action recognition［C］// Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Madison. Washington, DC: IEEE Computer Society, 2003: 1-7.
［3］AGGARWAL J K, CAI Q. Human motion analysis: a review［J］. Computer Vision and Image Understanding, 2005, 73(3): 428-440.
［4］THOMAS B M, ERIK G. A survey of computer vision-based human motion capture［J］. Computer Vision and Image Understanding, 2001,81(3): 231-268.
［5］WANG L, HU W M, TIE N T. Recent developments in human motion analysis［J］. Pattern Recognition, 2003, 36(3):585-601.
［6］WANG L, TAN T, NING H, et al. Silhouette analysis-based gait recognition for human identification［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(12):1505-1518.
［7］CAMPBELL L W, BECKER D A, AZARBAYEJANI A, et al. In-variant features for 3D gesture recognition［C］// Proceedings of the 1996 International Conference on Automatic Face and Gesture Recognition. Piscataway: IEEE, 1996:157-162.
［8］KALAL Z, MATAS J, MIKOLAJCZYK K. Online learning of robust object detectors during unstable tracking［C］// Proceedings of the 3rd On-line Learning for Computer Vision Workshop. Washington, DC: IEEE Computer Society, 2009:165-173.
［9］OJALA T, PIETIKAINEN M, MAENPAA T. Multi-resolution gray-scale and rotation invariant texture classification with local binary patterns［J］. International Journal of Computer Vision and Pattern Recognition, 2002, 24(7):971-987.
［10］RAHEJA J L, CHAUDHARV A, SIGNAL K. Tracking of fingertips and centers of palm using Kinect［C］// Proceedings of the 3rd International Conference on Computational Intelligence, Modeling and Simulation. Piscataway: IEEE, 2011:248-252.
［11］Open Kinect Organization. Imaging information for Kinect［EB/OL］.［2011-07-10］. http://openkinect.org/wiki/ImagingInformation.
［12］GU R, HE Y, JIN L. Design and implementation of Kinect skeletal tracking technology-based intelligent teaching control system［J］. Computer Application and Software,2013,30(6):276-280.(顾容,何益明,金丽燕. 基于Kinect骨骼追踪技术的智能教学控制系统的设计与实现［J］.计算机应用与软件, 2013,30(6):276-280.)

[1]	张佳慧李晓明张嘉祥. 强化形态感知的路面缺陷检测算法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[2]	杨建锋陈斌李雨轩. 基于点云重构的自监督点云异常检测方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[3]	薛振华李强黄超. 视觉大模型驱动的像素级图像异常检测模型[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[4]	蒋畅江向杰何旭颖. 面向机械臂抓取的双目视觉目标定位算法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[5]	边小勇胡其仁袁培洋. 多注意力对比学习的红外小目标检测[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[6]	李钟华钟庚辛范萍朱恒亮. 通过边界挖掘和背景引导的伪装目标检测[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[7]	吴松霖张广朝姚远彭博. 基于判别区域引导的多视图困难气道识别[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[8]	李强白少雄熊源袁薇. 基于视觉大模型隐私保护的监控图像定位[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[9]	薛雅丽徐忠敏刘世豪. 基于多级小波残差网络的重力数据去噪方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[10]	况世雄姚俊波陆佳炜王琪冰肖刚. 基于动态图卷积网络的电梯乘客异常行为数据增强方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[11]	康斌陈斌王俊杰李昱林赵军智咸伟志. 基于多粒度共享语义中心关联的文本到人物检索方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[12]	张庆杨凡方宇涵. 基于多模态信息融合的中文拼写纠错算法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[13]	王昊王金伟程鑫张家伟吴昊罗向阳马宾. 彩色图像JPEG重压缩取证综述[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[14]	王磊胡节彭博. 用于半监督火灾检测的分布自适应和动态课程伪标签框架[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[15]	刘晋文王磊马博董瑞杨雅婷艾合塔木江·艾合麦提王欣乐. 基于弱监督模态语义增强的多模态有害信息检测方法 [J]. 《计算机应用》唯一官方网站, 0, (): 0-0.