基于特征空间切分建模的变形手势跟踪算法

doi:10.13973/j.cnki.robot.170513

摘要
图/表
参考文献
相关文章 (2)

全文: PDF (907 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
张彦彬, 陈晓春. 基于特征空间切分建模的变形手势跟踪算法[J]. 机器人, 2018, 40(4): 401-412.DOI: 10.13973/j.cnki.robot.170513. ZHANG Yanbin, CHEN Xiaochun. The Deformed Gesture Tracking Algorithm Based on Feature SpaceSegmentation Modeling. ROBOT, 2018, 40(4): 401-412. DOI: 10.13973/j.cnki.robot.170513.

摘要为解决人机交互中手势形变和无规律运动带来的跟踪难题，提出了一种基于特征空间切分建模的非参数核密度估计算法来实现手势跟踪.首先，在检测模块中利用AdaBoost分类器检测图像中手势的存在，将检测到的手势位置信息传送给跟踪模块，该模块精确提取手势目标从而对其颜色建模.然后，利用目标的颜色模型对各帧图像进行后验概率密度估算，获取运动目标的概率密度图像，将其分解成手势运动区和同色干扰区.最后，对同色干扰区采用混合高斯建模来削弱同色目标的干扰.当目标丢失时启动再检测模块，并利用贝叶斯分类器与方差分类器实现手势目标重检.实验结果表明，该算法通过对特征空间切分建模以及不同分类器的级联解决了变形手势跟踪的同色干扰与再检测难题.该算法提高了跟踪的准确率（>81.5%），适合于非刚性物体做无规则运动的复杂场景.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张彦彬
	陈晓春

关键词 ：非参数核密度估计, 变形手势, 运动跟踪, 概率密度图, 同色干扰区, 再检测模块

Abstract：In order to solve the tracking problem caused by the deformation and irregular motion of gestures in the process of human-computer interaction, a nonparametric kernel density estimation algorithm based on the feature space segmentation modeling is proposed. Firstly, the AdaBoost classifier in the detection module is used to detect the presence of gestures in the image, and the gesture position information is sent to the tracking module, which accurately extracts the gesture target for color modeling. Then, the probability density image of motion target is obtained by estimating the posterior probability density of each frame image using the color model of the target, which is decomposed into the gesture motion area and the similar color interference region. Finally, the mixed Gauss model is used to weaken the interference of close color objects in the similar color region. The redetection module is started when the target is lost, and the gesture position can be detected by adopting the Bayesian classifier and the variance classifier. The experimental results show that the proposed method resolves the similar color interference and redetection problem in deformed gesture tracking by segmenting the feature space and cascading different classifiers. The proposed algorithm improves the tracking accuracy (>81.5%) and is suitable for complex scenes involving the irregular motion of nonrigid objects, and has a high tracking accuracy.

Key words： nonparametric kernel density estimation deformed gesture motion tracking probability density image similar color interference region redetection module

收稿日期: 2017-08-30 录用日期: 2018-02-13

TP319

基金资助:深圳市基础研究项目（JCYJ20160429170032960）；广东省教育部产学研结合计划（2012BO91100495）

通讯作者: 陈晓春,chenxc@tsinghua-sz.org E-mail: chenxc@tsinghua-sz.org

作者简介: 张彦彬(1990-),男,硕士生.研究领域:机器视觉和模式识别.
陈晓春(1972-),男,博士,高级工程师.研究领域:无线通信,人机交互,多媒体信息处理及其相关集成电路设计等.

链接本文:

https://robot.sia.cn/CN/10.13973/j.cnki.robot.170513 或 https://robot.sia.cn/CN/Y2018/V40/I4/401

[1] Mei K, Xu L, Li B, et al. A real-time hand detection system based on multi-feature[J]. Neurocomputing, 2015, 158:184-193.
[2] Yang C J, Duraiswami R, Davis L. Efficient mean-shift tracking via a new similarity measure[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, USA:IEEE, 2005:176-183.
[3] Chen C Y, Zhang M M, Qiu K J, et al. Real-time robust hand tracking based on Camshift and motion velocity[C]//5th International Conference on Digital Home. Piscataway, USA:IEEE, 2014:20-24.
[4] Hernandez-Belmonte U H, Ayala-Ramirez V. Real-time hand posture recognition for human-robot interaction tasks[J]. Sensors, 2016, 16(1):36.
[5] Xu W, Lee E J. A novel method for hand posture recognition based on depth information descriptor[J]. KSⅡ Transactions on Internet & Information Systems, 2015, 9(2):763-774.
[6] Chiang C T, Tseng P H, Feng K T. Hybrid unified Kalman tracking algorithms for heterogeneous wireless location systems[J]. IEEE Transactions on Vehicular Technology, 2012, 61(2):702-715.
[7] Huang H S, Zhou Y L, Chen P J, et al. Robust hand tracking with posture recognition via online learning[C]//IEEE International Conference on Cloud Computing and Intelligence Systems. Piscataway, USA:IEEE, 2014:65-70.
[8] Yang C J, Jang Y J, Beh J H. Gesture recognition using depth-based hand tracking for contactless controller application[C]//IEEE International Conference on Consumer Electronics. Piscataway, USA:IEEE, 2012:297-298.
[9] 严权峰,王岳斌,白天,等. 基于压缩感知的实时手势检测和跟踪算法[J]. 计算机工程与应用,2016,52(20):182-187.Yan Q F, Wang Y B, Bai T, et al. Real-time gesture detection and tracking algorithm based on compressive sensing[J]. Computer Engineering and Applications, 2016, 52(20):182-187.
[10] 侯荣波,康文雄,房育勋,等. 基于时间上下文跟踪-学习-检测的指尖跟踪方法[J]. 计算机应用,2016,36(5):1371-1377.Hou R B, Kang W X, Fang Y X, et al. Fingertip tracking method based temporal context tracking-learning-detection[J]. Journal of Computer Applications, 2016, 36(5):1371-1377.
[11] 左军毅,梁彦,潘泉,等. 基于多个颜色分布模型的Cam-shift跟踪算法[J]. 自动化学报,2008,34(7):736-742.Zuo J Y, Liang Y, Pan Q, et al. Camshift tracker based on multiple color distribution models[J]. Acta Automatica Sinica, 2008, 34(7):736-742.
[12] Cheng Y. Mean-shift, mode seeking, and clustering[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1995, 17(8):790-799.
[13] 关然,徐向民,罗雅愉,等. 基于计算机视觉的手势检测识别技术[J]. 计算机应用与软件,2013,30(1):155-159.Guan R, Xu X M, Luo Y Y, et al. A computer vision-based gesture detection and recognition technique[J]. Computer Applications and Software, 2013, 30(1):155-159.
[14] Overett G, Petersson L. Large scale sign detection using HOG feature variants[C]//Intelligent Vehicles Symposium. Piscataway, USA:IEEE, 2011:326-331.
[15] Freund Y, Schapire R E. A desicion-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer & System Sciences, 1997, 55(1):119-139.
[16] Weng L, Preneel B. A secure perceptual Hash algorithm for image content authentication[M]//12th IFIP TC6/TC11 International Conference on Communications and Multimedia Security. Berlin, Germany:Springer, 2011:108-121.
[17] Li K, Ran Y, Qin Q. Object-oriented port detection based on mean shift segmentation[C]//International Conference on Electrical and Control Engineering. Piscataway, USA:IEEE, 2010:1399-1402.
[18] Zivkovic Z. Improved adaptive Gaussian mixture model for background subtraction[C]//17th International Conference on Pattern Recognition. Piscataway, USA:IEEE, 2004:28-31.
[19] Sebe N, Cohen I, Huang T S, et al. Skin detection:A Bayesiannetwork approach[C]//International Conference on Pattern Re-cognition. Piscataway, USA:IEEE, 2004:903-906.
[20] 齐丽娜,张博,王战凯. 最大类间方差法在图像处理中的应用[J]. 无线电工程,2006,36(7):25-26.Qi L N, Zhang B, Wang Z K. Application of the Otsu method in image processing[J]. Radio Engineering, 2006, 36(7):25-26.
[21] Kalal Z, Mikolajczyk K, Matas J. Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2012, 34(7):1409-1422.
[22] 邱迪. 基于HSV与YCrCb颜色空间进行肤色检测的研究[J]. 电脑编程技巧与维护,2012(10):74-75.Qiu D. Research about skin color detection base on HSV and YCrCb color space[J]. Computer Programming Skills & Maintenance, 2012(10):74-75.
[23] Chai D, Bouzerdoum A. A Bayesian approach to skin color classification in YCbCr color space[C]//TENCON 2000. Piscataway, USA:IEEE, 2000:421-424.
[24] Bradski G R. Computer vision face tracking for use in a perceptual user interface[J]. Intel Technology Journal, 1998, Q2(Q2):214-219.
[25] 覃跃虎,支琤,徐奕. 基于三维直方图的改进Camshift目标跟踪算法[J]. 现代电子技术,2014,37(2):29-33.Qin Y H, Zhi C, Xu Y. Improved Camshift objects tracking algorithm based on three-dimensional histogram[J]. Modern Electronics Technique, 2014, 37(2):29-33.
[26] Yao M H, Gu Q L, Wang X B, et al. A novel hand gesture tracking algorithm fusing Camshift and particle filter[C]//Inter-national Conference on Artificial Intelligence and Industrial Engineering. Paris, France:Atlantis Press, 2015:37-40.
[27] Yu Y S, Bi S, Mo Y Y, et al. Real-time gesture recognition system based on Camshift algorithm and Haar-like feature[C]//IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems. Piscataway, USA:IEEE, 2016:337-342.
[28] Soni U, Trivedi A, Roberts N. Real-time hand tracking using integrated optical flow and Camshift algorithm[C]//2016 Second International Conference on Research in Computational Intelligence and Communication Networks. Piscataway, USA:IEEE, 2017:135-140.
[29] 刘亚伟,李小民. 基于BRISK和CamShift的鲁棒目标跟踪研究[J]. 电光与控制,2017,24(3):41-45.Liu Y W, Li X M. A robust target tracking method based on BRISK and CamShift[J]. Electronics Optics & Control, 2017, 24(3):41-45.