Parallel iterative median-type algorithms for image processing

151. Hong Tang, B. B. Zhou, I. D. G. Macleod and R. P. Brent, A class of parallel iterative median-type algorithms for efficient image processing, Proceedings of the International Conference on Systems, Control and Information - Methodologies and Applications (ICSCI), Wuhan, China, 1994, 21-28.


This paper proposes a class of parallel iterative median-type filters to restore signals degraded by Gaussian noise.  An efficient implementation of these filters on distributed memory MIMD machines is outlined.  Iterative median-type algorithms have been implemented with high efficiency (greater than 90%) on the Fujitsu AP1000.  We conclude that algorithms of this type are very suitable for parallel implementation on distributed memory machines.


For related papers, see [149, 152].

Go to next publication

Return to Richard Brent's index page