Adaptive kernel-bandwidth object tracking based on Mean-shift algorithm

With the wide application and development of Mean Shift algorithm, the shortages of the classical algorithm have been exposed. Among them, there is a defect that the kernel-bandwidth of the traditional object tracking algorithm based on Mean-Shift is fixed, which is very easy to cause the failure of the object tracking. To overcome this shortage, this paper puts forward an object tracking method that combines the canny operator with the Mean Shift algorithm. Firstly, this algorithm uses the canny edge detection to determine the change tendency of the object. Then proper ratio increment is used to adjust the Kernel-bandwidth of the Mean Shift. Finally, the object can be located accurately with appropriate kernel-bandwidth. The experimental results show that the improved algorithm can improves the tracking stability of the Mean Shift algorithm effectively when the size of the tracking object is changing and it is adaptive to the changing size of the tracking object.

[1]  Huang Qi.  Target Scale Adaptive Mean Shift Tracking Algorithm , 2011 .

[2]  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..

[3]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Dorin Comaniciu,et al.  Robust analysis of feature spaces: color image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  Larry D. Hostetler,et al.  The estimation of the gradient of a density function, with applications in pattern recognition , 1975, IEEE Trans. Inf. Theory.

[6]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Alper Yilmaz,et al.  Object Tracking by Asymmetric Kernel Mean Shift with Automatic Scale and Orientation Selection , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Dorin Comaniciu,et al.  Mean shift analysis and applications , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[9]  Yizong Cheng,et al.  Mean Shift, Mode Seeking, and Clustering , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Mubarak Shah,et al.  Target tracking in airborne forward looking infrared imagery , 2003, Image Vis. Comput..