DocumentCode :
2882318
Title :
A new method for moving object extraction automatically
Author :
Xiaoyan, Zhang ; Rongchun, Zhao ; Zhiqiang, Ma
Author_Institution :
Coll. of Comput., Northwestern Polytech Univ., Xi´´an, China
Volume :
2
fYear :
2005
fDate :
12-14 Oct. 2005
Firstpage :
1460
Lastpage :
1463
Abstract :
A novel algorithm for moving object extraction is proposed in this paper. First of all, the algorithm applies the inter-frame change detection and Canny edge detection in wavelet domain to obtain the accurate difference edge map. Then, through combination with the current frame edge map, background edge map and previous frame´s moving edge, the current frame´s edge map named moving edge can be tracked. At last, the moving edge is used to extract video object planes (VOP) by a simple filling technique. The proposed algorithm is robust to the entire motion and local deformation of object. Experimental results for several video sequences demonstrate the effectiveness of the proposed approach.
Keywords :
edge detection; image motion analysis; image sequences; object detection; video signal processing; wavelet transforms; Canny edge detection; background edge map; difference edge map; frame edge map; frames moving edge; interframe change detection; moving object extraction; simple filling technique; video object planes extraction; video sequences; wavelet domain; Change detection algorithms; Humans; Image edge detection; Low pass filters; Object segmentation; Partitioning algorithms; Video sequences; Wavelet analysis; Wavelet domain; Wavelet transforms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Communications and Information Technology, 2005. ISCIT 2005. IEEE International Symposium on
Print_ISBN :
0-7803-9538-7
Type :
conf
DOI :
10.1109/ISCIT.2005.1567146
Filename :
1567146
Link To Document :
بازگشت