Parallel median filtering using cellular neural networks
Saeid Sadeghi-Emamchaie, GRAHAM A. JULLIEN, W.C. Miller · 2002
In this paper we describe methods of designing cellular neural networks (CNNs) for parallel median filtering of noisy images. The paper contains two contributions. The first is a cellular neural network structure, with linear templates, for median filtering of binary images. The second is a new technique for designing cellular neural networks for the special, but important, case of gray-scale separable median filters with a window size of 3/spl times/3.