ADAPTIVE MULTISTAGE WEIGHTED ORDER STATISTIC FILTERS BASED ON THE BACKPROPAGATION ALGORITHM

As a concise representation of stack filters, multistage weighted-orde r statistic (MWOS) filters are introduced in this paper, which corresp ond to multistage threshold logic gates or multilayer perceptrons in t he binary domain. Two adaptive algorithms are derived for finding opti mal MWOS filters under the MAE criterion and the MSE criterion, respec tively. Experimental results from image enhancement are provided to co mpare the performance of adaptive MWOS filters and adaptive stack filt ers.