Automated human tracking using advanced mean shift algorithm

For tracking human or any kind of object, colour feature based mean shift technique is widely used. This technique uses Bhattacharya coefficient to locate the object based on the maximisation of the similarity function between object model and candidate model. Traditional mean shift algorithm fails when the object having large motion, occlusion, corrupted frames etc. In addition to that, the technique is not automatic to initiate the tracking. To overcome all these problems, this paper proposes a technique which uses three additional modules to the traditional method to make it more efficient. The proposed modules uses human detection by modelling through star skeletonization, followed by block search algorithm and occlusion handling. Block search algorithm helps to supply an overlapping area to candidate model to continue the track when tracking fails due to fast motion. Occlusion handling will help to initiate the tracking after prolonged period of occlusion. The proposed method has been tested on real time data and it outperforms the conventional method effectively to overcome the mentioned problems up to large extent.

[1]  Luping Wang,et al.  An adaptive tracking window based on mean-shift target tracking algorithm , 2013, 2013 Chinese Automation Congress.

[2]  Amir Hooshang Mazinan,et al.  Applying mean shift, motion information and Kalman filtering approaches to object tracking. , 2012, ISA transactions.

[3]  Hou Zhi-qiang,et al.  Mean-Shift Tracking Algorithm with Improved Background-Weighted Histogram , 2014, 2014 Fifth International Conference on Intelligent Systems Design and Engineering Applications.

[4]  Zixing Cai,et al.  A Robust Object Tracking Approach using Mean Shift , 2007, Third International Conference on Natural Computation (ICNC 2007).

[5]  N. Zarka,et al.  Real-Time Human Motion Detection and Tracking , 2008, 2008 3rd International Conference on Information and Communication Technologies: From Theory to Applications.

[6]  Kenji Terada,et al.  A framework for Human tracking using Kalman filter and fast mean shift algorithms , 2009, 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops.

[7]  Dorin Comaniciu,et al.  Real-time tracking of non-rigid objects using mean shift , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[8]  David A. Clausi,et al.  Efficient Target Recovery Using STAGE for Mean-shift Tracking , 2009, 2009 Canadian Conference on Computer and Robot Vision.

[9]  Xinbo Gao,et al.  Incremental tensor biased discriminant analysis: A new color-based visual tracking method , 2010, Neurocomputing.

[10]  R. Venkatesh Babu,et al.  Robust Object Tracking using Local Kernels and Background Information , 2007, 2007 IEEE International Conference on Image Processing.

[11]  Zhenyang Wu,et al.  A mean shift algorithm based on modified Parzen window for small target tracking , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[12]  B. Kröse,et al.  An EM-like algorithm for color-histogram-based object tracking , 2004, CVPR 2004.

[13]  Hélène Laurent,et al.  Review and evaluation of commonly-implemented background subtraction algorithms , 2008, 2008 19th International Conference on Pattern Recognition.