Title :
Restoration of noisy images by the adaptive average differential iterative and the improved median filters
Author_Institution :
Dept. of Electron. Eng., La Trobe Univ., Bundoora, Vic., Australia
Abstract :
In the literature there are many methods to restore noisy images. An adaptive differential iterative filter (AADIF) and an improved median filter (IMF) which suppress impulsive noise and Gaussian noise while preserving the image edges and details successfully, are analysed and evaluated in terms of their filtering performances under practical noise conditions. The signal-to-noise ratios in this experiment are from 0 dB to 17 dB. Besides the description of their algorithms, a great deal of experimental data are shown to support their efficacy. It is concluded that these two filters are highly competitive.<>
Keywords :
edge detection; image reconstruction; Gaussian noise; adaptive average differential iterative filter; image edges; improved median filters; impulsive noise; noisy image restoration; signal-to-noise ratios; Adaptive filters; Additive noise; Cellular neural networks; Filtering; Fuzzy neural networks; Image restoration; Noise level; Pixel; Signal to noise ratio; Testing;
Conference_Titel :
TENCON '93. Proceedings. Computer, Communication, Control and Power Engineering.1993 IEEE Region 10 Conference on
Conference_Location :
Beijing, China
Print_ISBN :
0-7803-1233-3
DOI :
10.1109/TENCON.1993.320196
