Template Matching Based on Polar Coordinate

Abstract:

A new template matching algorithm based on polar coordinate is proposed to improve the performance of the tracking system. The shape of the matching template is round, and the pixels in the template and the matching area are arranged into circle, which can ensure the rotation invariance of the method. Some differential matching information is added into the matching criterion function, which makes the method have higher recognition precision. In the long time target tracking process, template updating operation is adopted to avoid losing the target. Simulation results prove that the method can be applied in the TV tracking system. Using the same controller, the method has better tracking performance.

• Periodical: Advanced Materials Research (Volumes 524-527)
• Pages: 3774-3777
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.524-527.3774
• Online since: May 2012
• Authors: Xu Juan Miao, Xiao Fei Li
• Keywords: Polar Coordinate, Target Recognition, Template Matching, Tracking System

References

[1] K. C. Lee, J. Ho, M. H. Yang and D. Kriegman: Computer Vision and Image Understanding. Vol. 99 (2005), p.303

[2] S. S. Ge, Y. Yang and T. H. Lee: Image and Vision Computing. Vol. 26 (2008), p.1607

[3] J. C. Tai, S. T. Tseng, C. P. Lin and K. T. Song: Image and Vision Computing. Vol. 22 (2004), p.485

[4] H. C. Kuo and L. J. Wu: Journal of Materials Processing Technology. Vol. 120 (2002), p.169

[5] S. Kang and S. W. Lee: Pattern Recognition. Vol. 35 (2002), p. (2031)

[6] J. Shaik and K. M. Iftekharuddin: Optics & Laser Technology. Vol. 41 (2009), p.832

[7] H. Li, H. B. Duan and X. Y. Zhang: Pattern Recognition Letters. Vol. 31 (2010), p.1825

[8] C. L. Li and K. C. Hui: Computers & Graphics. Vol. 24 (2000), p.569

[9] S. Mattoccia, F. Tombari and L. D. Stefano: Pattern Recognition Letters. Vol. 32 (2011), p.694

[10] I. Kumazawa: Pattern Recognition Letters. Vol. 21 (2000), p.661