Parallel iterative noise filters for real-time image processing

152. Hong Tang, B. B. Zhou, I. D. G. Macleod, R. P. Brent and Wei Sun, Comparisons of parallel iterative noise filters for real-time image processing, Proc. Fifth International Conference on Signal Processing Applications and Technology (ICSPAT), Dallas, Texas, 1994, Vol. 2, 1015-1020.

Abstract

In this paper, we investigate two parallel iterative noise filters for restoration of noise-degraded images.  These filters are analysed and evaluated in terms of their filtering performance under practical noise conditions, with signal-to-noise ratios ranging from 0dB to 10dB.  The parallel iterative noise filters have been implemented with high efficiency on a Fujitsu AP1000 mesh-connected array processor.  The algorithms and their method of parallel implementation are presented together with the results of extensive experiments which demonstrate the effectiveness and speed of the filters.  We conclude that these two forms of parallel filter are promising candidates for real-time image restoration.

Comments

For related papers, see [149, 151].

Go to next publication

Return to Richard Brent's index page