Presentation 2014-03-07
Design of Ll filters based on local shape clustering of images
Masanori SAITO, Mitsuhiko MEGURO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a new design method of a Ll-filter based on local characteristics in images. It is not preferable for a filter to use the same filter coefficients whole image which tends to be a non-stationary signal. Therefore, at first, we divide each local image region into each class which belongs to the same shape of intensity value in local region in image. For the reason, we divide the local image region to each class by K-means and local SNR information. Secondly, we design each Ll-filter belongs to each class by LMS algorithms. Lastly, we use a prefer Ll filter in each local image characteristics.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) non-linear filter / Ll filter / clustering
Paper # SIS2013-69
Date of Issue

Conference Information
Committee SIS
Conference Date 2014/2/27(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 Ll filters based on local shape clustering of images
Sub Title (in English)
Keyword(1) non-linear filter
Keyword(2) Ll filter
Keyword(3) clustering
1st Author's Name Masanori SAITO
1st Author's Affiliation Depertment of Mathematical Information Engineering, Cllege of Industrial Technology, Nihon University()
2nd Author's Name Mitsuhiko MEGURO
2nd Author's Affiliation Depertment of Mathematical Information Engineering, Cllege of Industrial Technology, Nihon University
Date 2014-03-07
Paper # SIS2013-69
Volume (vol) vol.113
Number (no) 467
Page pp.pp.-
#Pages 5
Date of Issue