Title :
Optical Flow Estimation Algorithm Time-Based Multi Correlation Matching
Author :
Xu, Jing ; Yang, Huamin
Author_Institution :
Sch. of Comput. Sci. & Technol., Changchun Univ. of Sci. & Technol., Changchun, China
Abstract :
Moving estimation is a hot issue in computer vision research. This paper proposes the use of prediction vector and alternate vectors to estimate optical flow, using time-based multi correlation method to compute prediction vector, based on the credibility evaluation function to determine the motion vector. Finally, this paper presents the algorithm´ s application case in moving target analysis.
Keywords :
computer vision; image sequences; motion estimation; alternate vectors; computer vision; credibility evaluation function; moving estimation; moving target analysis; optical flow estimation algorithm; prediction vector; time-based multicorrelation matching; Application software; Computer science; Computer vision; Correlation; Distributed computing; Image motion analysis; Industrial engineering; Information management; Innovation management; Optical computing; alternate vectors; computer vision; optical flow; prediction vector;
Conference_Titel :
Information Management, Innovation Management and Industrial Engineering, 2009 International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-0-7695-3876-1
DOI :
10.1109/ICIII.2009.432