Title :
The output distribution of fuzzy weighted median filters
Author :
Nie, Yao ; Barner, Kenneth E.
Author_Institution :
Dept. of Electr. & Comput. Eng., Delaware Univ., Newark, DE, USA
Abstract :
Fuzzy weighted median (FWM) filters are developed by utilizing concepts of fuzzy ordering in the weighted median (WM) filters framework and have great advantages over their crisp counterparts. In this paper, two important properties of fuzzy ordering - the element invariant property and order invariant property - are investigated, which shall serve as the theoretical justification of FWM filters. Based on these two properties, we derive the analytical expressions for the output distribution of FWM filters, which provide systematic methods to study the statistical properties of FWM filters.
Keywords :
filtering theory; fuzzy set theory; median filters; element invariant property; fuzzy ordering; fuzzy weighted median filters; nonlinear filtering; order invariant property; output distribution; statistical properties; Design methodology; Filtering algorithms; Genetic mutations; Image denoising; Information filtering; Information filters; Signal processing; Signal processing algorithms; Statistics; Strontium;
Conference_Titel :
Circuits and Systems, 2002. MWSCAS-2002. The 2002 45th Midwest Symposium on
Print_ISBN :
0-7803-7523-8
DOI :
10.1109/MWSCAS.2002.1186967
