Presentation 2012-03-01
Design of Optimal Noise Detector for Impulsive Noise Removal By Switching Vector Median Filters
Makoto KAMIYA, Yoshiko HANADA, Mitsuji MUNEYASU,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Vector Median Filters (VMF) that consider the correlation among the color components are often used for impulse noise removal. In addition Switching VMFs (SWVMFs) have been proposed to improve the preservation of pixel values in noncorrupt pixels. The filtering performance of SWVMFs depends on not only the accuracy in the filtering step but also the accuracy in the noise detection step, therefore, the estimation of a suitable window shape for the noise detection step in SWVMFs is one of important problems. In this study, for color textures, we formulate the design of the noise detector for SWVMFs as an optimization problem, and estimate optimal shapes directly from degraded images. Numerical experiments show that designed noise detectors outperform general filtering shapes.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Impulsive Noise / Switching Vector Median Filters / Window shapes / Genetic Algorithm
Paper # SIS2011-63
Date of Issue

Conference Information
Committee SIS
Conference Date 2012/2/23(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Design of Optimal Noise Detector for Impulsive Noise Removal By Switching Vector Median Filters
Sub Title (in English)
Keyword(1) Impulsive Noise
Keyword(2) Switching Vector Median Filters
Keyword(3) Window shapes
Keyword(4) Genetic Algorithm
1st Author's Name Makoto KAMIYA
1st Author's Affiliation Graduate School of Science and Engineering, kansai University()
2nd Author's Name Yoshiko HANADA
2nd Author's Affiliation Faculty of Engineering Science, kansai University
3rd Author's Name Mitsuji MUNEYASU
3rd Author's Affiliation Faculty of Engineering Science, kansai University
Date 2012-03-01
Paper # SIS2011-63
Volume (vol) vol.111
Number (no) 457
Page pp.pp.-
#Pages 6
Date of Issue