DocumentCode
306410
Title
Biologically motivated neural computing in early vision processing
Author
Guan, L. ; Anderson, J.A. ; Sutton, J.P.
Author_Institution
Dept. of Electr. Eng., Sydney Univ., NSW, Australia
Volume
2
fYear
1996
fDate
14-17 Oct 1996
Firstpage
1138
Abstract
We introduce a network of networks (NoN) model to solve image processing problems in early vision. The method is motivated by the fact that natural image formation is a local process, and thus processing can be accomplished by a globally coordinated, local parallel processing structure, readily implemented by the hierarchical cluster architecture of the NoN model. The modeling is very powerful in that it achieves high quality adaptive processing, and virtually eliminates the computational difference between inhomogeneous and homogeneous conditions. Computer simulations show that this method is able to provide fast, quality image processing in early vision
Keywords
computer vision; neural nets; parallel processing; quadratic programming; adaptive processing; early vision processing; hierarchical cluster architecture; image formation; image processing; image regularization; network of networks model; neural computing; parallel processing; quadratic programming; Biology computing; Computer vision; Degradation; Image processing; Image sensors; Neural networks; Optimization methods; Parallel processing; Pixel; Sensor arrays;
fLanguage
English
Publisher
ieee
Conference_Titel
Systems, Man, and Cybernetics, 1996., IEEE International Conference on
Conference_Location
Beijing
ISSN
1062-922X
Print_ISBN
0-7803-3280-6
Type
conf
DOI
10.1109/ICSMC.1996.571244
Filename
571244
Link To Document