Title :
Image restoration based template matching for multichannel restoration of image sequences
Author :
Choi, Mun Gi ; Galatsanos, Nikolas P. ; Schonfeld, Dan
Author_Institution :
Dept. of Electr. & Comput. Eng., Illinois Inst. of Technol., Chicago, IL, USA
Abstract :
It has been demonstrated that multichannel restoration of motion compensated images sequences is very effective when good estimates of the displacement vector field (DVF) are available. In this paper two new stochastic algorithms for template-matching based DVF estimation are proposed. These algorithms are based on the relationship of the template-matching and the image restoration problems. Experiments are shown where the valve of the proposed DVF estimation algorithms is demonstrated.
Keywords :
filtering theory; image matching; image restoration; image sequences; maximum likelihood estimation; motion compensation; stochastic processes; displacement vector field estimation; image restoration; image sequences; maximum likelihood estimation; motion compensated images; multichannel restoration; stochastic algorithms; template matching; Computer science; Degradation; Image coding; Image restoration; Image sequences; Maximum likelihood estimation; Motion estimation; Object recognition; Parameter estimation; Stochastic processes;
Conference_Titel :
Signals, Systems and Computers, 1996. Conference Record of the Thirtieth Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-8186-7646-9
DOI :
10.1109/ACSSC.1996.600835
