Electrical and Computer Engineering, Temple University, Philadelp hia 19122, USA
Abstract-A snake algorithm has been known that it has a strong point in extracting the exact contour of an object. But it is apt to be influen ced by scattered edges around the control points. Since the shape of a moving ob ject in 2D image changes a lot due to its rotation and translation in the 3D spa ce, the conventional algorithm that takes into account slowly moving objects can not provide an appropriate solution. To utilize the advantages of the snake algo rithm while minimizing the drawbacks, this paper proposes the area variation bas ed color snake algorithm for moving object tracking. The proposed algorithm incl udes a new energy term which is used for preserving the shape of an object betwe en two consecutive images. The proposed one can also segment precisely interesti ng objects on complex image since it is based on color information. Experiment r esults show that the proposed algorithm is very effective in various environments.
Key words-color snake algorithm; area variation; moving object tracking; snake energy; segmentation
Manuscript Number: 1674-8042(2010)01-0046-04
dio: 10.3969/j.issn.1674-8042.2010.01.09
References
[1]M. Kass, A. Witkin, D. Terzopoulos, 1987. Snakes: active contour mod els. International Journal of Computer Vision, 1(4): 321-331 .
[2]C. Xu, J. L. Prince, 1997. Gradient Vector Flow: a New External Forc e for Active Contours. Proc. IEEE Conf. on Comp. Vis. Patt. Recog. (CVPR), p. 66 -71.
[3]A. Koschan, 2003. Color active shape models for tracking non-rigid objects. Pattern Recognition Letters, 24: 1751-1765.
[4]L. Pi, J. Fan, C. Shen, 2007. Color image segmentation for objects o f interest with modified geodesic active contour method. Journal of Ma thematical Imaging, 27: 51-57.
[5]C. Xu, A. Yezzi, J. L. Prince, 2000. On the Relationship between Par ametric and Geometric Active Contours. Asilomar Conference on Signals, Systems, and Computers, p. 483-489.
[full text view]
