A new tunable fuzzy filter for image smoothing
H.S. Kam, Wooi-Haw Tan, Madasu Hanmandlu · 2004
Different applications sometimes require varying degrees of smoothing enhancement so as not to tradeoff important image detail with increased smoothing. A sigmoid membership function is proposed in this paper to model the image information in the spatial domain. This function acts as a smoothing intensification operator or a tunable fuzzy filter. With a proper choice of two parameters 't' and 'a', the filter strength can be tuned for removal of noise in gray images. A visible improvement in the smoothness of images is observed, and the output of filter is also compared with those of other standard smoothing methods.