Aimed at the defect of classic Mean-shift tracking algorithm which is vulnerable to similar background inference for using single color feature, an improved color and texture features combined Mean-shift tracking algorithm is presented. Feature joint similarity was introduced for the first time. By maximizing the integrated similarity using Mean-shift algorithm the center of target in the new frame can be obtained accurately and rapidly. Experiments show that the proposed algorithm provides more reliable performance while satisfying the real-time requirements of general target tracking tasks.
[1] DRUMMOND T, CIPOLLA R. Realtime visual tracking of complex structures [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 4(7): 932-946.
[2] COMANICIU D, RAMESH V, MEER P. Kernelbased object tracking [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577.
[3] LIU T L, CHEN H T. Realtime tracking using trustregion methods [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(3): 397-402.
[4] COLLINS R T. Meanshift blob tracking through scale space [C]∥ Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Madison Wisconsin: IEEE CS, 2003, 2: 234-240.
[5] ZIVKOVIC Z, KROSE B. An EMlike algorithm for colorhistogrambased object tracking [C]∥ Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washirgton D. C.: IEEE CS, 2004: 798-803.
[6] STERN H, EFROS B. Adaptive color space switching for face tracking in multicolor lighting environment [C]∥ Proceedings of IEEE Conference on Automatice Face and Gesture Recognition. Washirgton D. C.: IEEE CS,2002: 249-254.
[7] COLLINS R T, LIU Y. Online selection of discriminative tracking features [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10): 1631-1643.
[8] POLAT E, OZDEN. Anonparametric adaptive tracking algorithm based on multiple feature distributions [J]. IEEE Transactions on Multimedia, 2006, 8: 1156-1163.
[9] WANG J Q, YAGI Y. Integrating shape and color features for adaptive realtime object tracking [C]∥ Proceedings of IEEE Conference on Robotics and Biomimetics. Kunming: IEEE RAS, 2006: 1-6.
[10] TUCERYAN M, JAIN A K. Texture analysis, handbook pattern recognition and computer vision [M]. Singapore: World Scientific, 1993: 235-276.
[11] STOLLNITZ E J, DEROSE T D, SALESIN D H. Wavelets for computer graphics: A primer part 1 [J] IEEE Computer Graphics and Application, 1995, 15: 76-84.
[12] MANJUNATH B S, MA W Y. Texture features for browsing andretrieval of large imagedata [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(8): 837-842.
[13] KARASARIDIS A, SIMONCELLI E. A filter design technique for steerable pyramid image transforms [C]∥ Proceedings of International Conference on Acoustics Speech and Signal Processing. Atlanta, Georgia: IEEE SPS, 1996: 2387-2390.