Title :
Optical flow field estimation of nature scene images
Author :
Long, Ma ; Lu-ping, Wang
Author_Institution :
ATR Key Lab., Nat. Univ. of Defense Technol., Changsha, China
Abstract :
Covered-uncovered problem and noise interference are key problems in optical flow estimation of nature scene images. In this paper, the visible scene areas are labeled by visible matrix and the distribution of noise is approximately described with normal probability distribution. The arithmetic expressions of optical flow field in visible scene areas are deduced in the Bayes framework. The results of simulation experiments show that the method is effective.
Keywords :
Bayes methods; image sequences; interference; matrix algebra; noise; statistical distributions; Bayes framework; covered-uncovered problem; nature scene images; noise distribution; noise interference; normal probability distribution; optical flow estimation; optical flow field estimation; visible matrix; Image resolution; Optical imaging; Covered-uncovered; Image processing; Noise interference; Optical flow field;
Conference_Titel :
Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6539-2
DOI :
10.1109/ICACTE.2010.5579609
