Title :
Parallel iterations for recursive median filter
Author :
Burian, Adrian ; Takala, Jamo ; Topa, Marina Dana
Author_Institution :
Inst. of Digital & Comput. Syst., Tampere Univ. of Technol., Finland
Abstract :
In this paper we study the possibilities to increase the speed of convergence of the recursive median filter (RMF) by using parallel iterations. The median cost function is used for this purpose. The convergence behavior and the speed-up of the proposed filters are studied. The effects of the parallelization of the filtering process over the performances, as well as its efficiency, are also considered.
Keywords :
convergence of numerical methods; image processing; iterative methods; median filters; recursive filters; smoothing methods; Lena image; convergence speed; filtering process parallelization; median cost function; parallel iterations; recursive median filter; smoothing capability; Attenuation; Binary sequences; Concurrent computing; Convergence; Cost function; Digital filters; Filtering; Nonlinear filters; Power smoothing; Signal processing;
Conference_Titel :
Circuits and Systems, 2003. ISCAS '03. Proceedings of the 2003 International Symposium on
Print_ISBN :
0-7803-7761-3
DOI :
10.1109/ISCAS.2003.1205926
