Vector median-rational hybrid filters for multichannel imageprocessing

摘要

In this letter, a new class of nonlinear filters called vectornmedian-rational hybrid filters (VMRHFs) for multispectral imagenprocessing is introduced and applied to the color image filteringnproblem. These filters are based on rational functions (RFs) offering annumber of advantages. First, a rational function is a universalnapproximator and a good extrapolator. Second, it can be trained by anlinear adaptive algorithm. Third, it produces the best approximationn(w.r.t. a given cost function) for some specific functions. The outputnis the result of a vector rational operation over the output of threensubfilters, such as two vector median (VM) subfilters and one centernweighted vector median filter (CWVMF). These filters exhibit desirablenproperties, such as edge and details preservation and accuratenchromaticity estimation