DocumentCode :
2837074
Title :
An Efficient Method of Dynamic Texture Tracking Based on Increment Evolution
Author :
Quan, Hongyan ; Wang, Changbo
Author_Institution :
Inst. of Software Eng., East China Normal Univ., Shanghai, China
fYear :
2009
fDate :
11-13 Dec. 2009
Firstpage :
1
Lastpage :
4
Abstract :
Dynamic texture tracking is a hot spot problem of computer vision, video surveillance and other areas. This paper presents a dynamic texture real-time tracking method .We first put forward a dynamic texture segmentation method based on level set based. In the method we define the evolution function, use the re-initialization condition to punish in the evolution process and use the optical flow based terminal condition to limit the evolution in order to make the level set more efficient. It overcomes the shortcoming of reinitializing and slow evolution speed problem of level set in the existing methods. In the study of dynamic texture tracking, in order to realize real-time tracking, we use the relevance of adjacent frames to evolution. Experimental results show that the method of real-time tracking of dynamic texture is a very effective.
Keywords :
computer vision; image segmentation; image texture; real-time systems; video surveillance; adjacent frames evolution; computer vision; dynamic texture segmentation method; dynamic texture tracking; efficient method; evolution function; evolution process; evolution speed problem; hot spot problem; increment evolution; optical flow based terminal condition; real time tracking method; reinitialization condition; video surveillance; Computer graphics; Computer vision; Image motion analysis; Image segmentation; Level set; Nonlinear optics; Optical devices; Robots; Software engineering; Video surveillance;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Software Engineering, 2009. CiSE 2009. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-4507-3
Electronic_ISBN :
978-1-4244-4507-3
Type :
conf
DOI :
10.1109/CISE.2009.5364503
Filename :
5364503
Link To Document :
بازگشت