Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20


Journal of Information Science and Engineering, Vol. 24 No. 2, pp. 521-537 (March 2008)

New Fuzzy Object Segmentation Algorithm for Video Sequences*

Kuo-Liang Chung, Shih-Wei Yu, Hsueh-Ju Yeh, Yong-Huai Huang and Ta-Jen Yao
Department of Computer Science and Information Engineering
National Taiwan University of Science and Technology
Taipei, 106 Taiwan

A new fuzzy moving object segmentation algorithm for video sequences is presented in this paper. Our proposed efficient object segmentation algorithm consists of three steps, namely the spatial segmentation step, the temporal tracking step, and the step for identifying the moving object from the frame in a fuzzy way. Especially, our proposed algorithm can robustly distinguish the foreground part, which is a near stationary region but surrounded by some regions with moving variation, from the background. Using several different real video sequences, experimental results demonstrate that the object segmentation accuracy of our proposed fuzzy-based algorithm is encouraging when compared to the recently published object segmentation algorithms by Chien et al. and Kim et al.

Keywords: fuzzy approach, moving object segmentation, MPEG-4, spatial segmentation, temporal tracking

Full Text () Retrieve PDF document (200803_13.pdf)

Received March 17, 2006; revised September 8 & December 14, 2006; accepted January 16, 2007.
Communicated by Ja-Ling Wu.
*This work was partially supported by the National Science Council of Taiwan, R.O.C. under contract No. 95- 2221-E011-152.