A robust algorithm for tracking object under occlusion and illumination change

An adaptive threshold value (ATV) and two-orthogonal-orientation edge correlogram (TOEC) based algorithm is proposed for tracking moving object in real scenarios. The ATV is used to extract the object edges under illumination changes. To improve the object edges representation power, the TOEC encodes the edge orientation pair levels along two orthogonal directions explicitly. An entropy weighting-maximization scheme is presented to achieve the maximum likelihood estimation of the similar regions and scales. Experimental results show that the proposed approach is appealing with respect to the robustness in the scenarios of complex occlusions and illumination variations.

[1]  Isabelle Bloch,et al.  Visual tracking by fusing multiple cues with context-sensitive reliabilities , 2012, Pattern Recognit..

[2]  Jamshid Shanbehzadeh,et al.  Image retrieval based on shape similarity by edge orientation autocorrelogram , 2003, Pattern Recognit..

[3]  Robert T. Collins,et al.  Mean-shift blob tracking through scale space , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[4]  Qi Zhao,et al.  A motion observable representation using color correlogram and its applications to tracking , 2009, Comput. Vis. Image Underst..

[5]  Zhang Yu-jin,et al.  Edge-color-histogram and Kalman filter-based real-time object tracking , 2008 .

[6]  Pinit Kumhom,et al.  Robust ambiguous target handling for visual object tracking , 2010 .

[7]  Stanley T. Birchfield,et al.  Spatiograms versus histograms for region-based tracking , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[8]  Gang Yu,et al.  Illumination Invariant Object Tracking with Incremental Subspace Learning , 2009, 2009 Fifth International Conference on Image and Graphics.

[9]  Tony Lindeberg,et al.  Feature Detection with Automatic Scale Selection , 1998, International Journal of Computer Vision.

[10]  Shumin Fei,et al.  An occlusion tolerent method for multi-object tracking , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[11]  Paul Bao,et al.  Image retrieval based on multi-scale edge model , 2002, Proceedings. IEEE International Conference on Multimedia and Expo.

[12]  Tim Ellis,et al.  Tracking occluded objects using partial observation , 2003 .

[13]  Suk-Ho Lee,et al.  Scale-adaptive object tracking using color centroids , 2011 .

[14]  Dorin Comaniciu,et al.  Kernel-Based Object Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Nitin S. Ujgare,et al.  Kernel Based Object Tracking Using Mean Shift Method , 2012 .

[16]  Dewen Hu,et al.  Tracking objects using shape context matching , 2012, Neurocomputing.