DocumentCode :
2113953
Title :
An edge closing method of motion segmentation based on a hybrid field snake
Author :
Xiaoqing Wu ; Xiang Long ; Xiong Yang
Author_Institution :
Coll. of Inf. Sci. & Technol., Hainan Univ., Haikou, China
fYear :
2013
fDate :
23-25 July 2013
Firstpage :
1030
Lastpage :
1034
Abstract :
We proposed a motion segmentation method a few months ago. The problem of it is that the contour of the object segmented from the image sequence is always unclosed. This paper presents a method of the Gradient Vector Flow (GVF) field snake to close the contour. The initial curve of the GVF snake is rectangle, so that the curve can be used in the tracking and recognition applications. However, corners of initial curve are always located out of the GVF field. In order to evolve the initial curve correctly, a hybrid field snake method is proposed to solve this problem. In the final part of this paper, we give some experiments to test our method, and the result is satisfactory.
Keywords :
image motion analysis; image segmentation; image sequences; video signal processing; GVF field snake; edge closing method; gradient vector flow; hybrid field snake; image sequence; motion segmentation; object contour segmentation; recognition application; tracking application; Computer vision; Hidden Markov models; Image edge detection; Image segmentation; Motion segmentation; Shape; Tracking; GVF snake; edge closing; hybrid field; motion segmentation; shrinking field;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2013 10th International Conference on
Conference_Location :
Shenyang
Type :
conf
DOI :
10.1109/FSKD.2013.6816347
Filename :
6816347
Link To Document :
بازگشت