In order to cope with the complex variation of target appearance during visual tracking,a robust tracking algorithm based on feature matching of key regions is proposed.Firstly,it initializes the target model and obtains target candidate through filter prediction.Then,it extracts the key regions of target model and target candidate using adaptive marker-based watershed algorithm and describes them with multiple features.Finally,it matches the key regions to get the mapping from target model to target candidate and calculates the final tracking results to output and update the target model.The proposed algorithm is tested on the video database containing the appearance variation of scale,occlusion,rotation,illumination,pose,background clutters,and motion blur.The experimental results demonstrate that the proposed algorithm can well cope with the complex appearance variation,especially shows the robustness to the partial occlusion,illumination and background clutters.
[1] DComaniciu,et al.Kernel-based object tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2003,25(5):564-577.
[2] AAdam,et al.Robust fragment-based tracking using integral histogram[A].In:Proceedings of the 21st IEEE Conference on Computer Vision and Pattern Recognition[C].New York,USA,IEEE Press,2008.798-805.
[3] HGrabner,et al.Real-time tracking via online boosting[A].In:Proceedings of British Machine Vision Conference[C].Edinburgh,UK,BMVA Press,2006.47-56.
[4] BBabenko,et al.Robust object tracking with online multiple instance learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2011,33(8):1619-1632.
[5] RYao,et al.Part-based visual tracking with online latent structural learning[A].In:Proceedings of the 26th IEEE Conference on Computer Vision and Pattern Recognition[C].Portland,USA,IEEE Press,2013.2363-2370.
[6] 蔺海峰,等.基于SIFT 特征目标跟踪算法研究[J].自动化学报.2010,36(8):1024-1028. H F Lin,et al.Research on object tracking algorithm based on SIFT[J].Acta Automatica Sinica.2010,36(8):1024-1028.(in Chinese)
[7] 刘玉,等.一种基于SIFT 和KLT 相结合的特征点跟踪方法研究[J].宇航学报,2011,32(7):1618-1627. YLiu,et al.A feature point tracking method based on the combination of SIFT algorithm and KLT matching algorithm[J].Journal of Astronautics,2011,32(7):1618-1627.(in Chinese)
[8] BFeng,et al.Tracking object by combining particle filters and SIFT features[A].In:Proceedings of the 5th International Conference on Image and Graphics[C].Xi'an,China,IEEE Press,2009.527-532.
[9] HZhou,et al.Object tracking using SIFT features and mean shift[J].Computer Vision and Image Understanding.2009,113(3):345-352.
[10] 余旺盛,等.基于标记分水岭和区域合并的彩色图像分割[J].电子学报,2011,39(5):1007-1012. W S Yu,et al.Color image segmentation based on marked-watershed and region-merger[J].Acta Electronica Sinica,2011,39(5):1007-1012.(in Chinese)
2014, Vol. 42 
