Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.kei.com!simtel!news.sprintlink.net!demon!btnet!uunet!nntp.hk.super.net!news.ust.hk!william
From: william@cs.ust.hk (Cheung Kwok Wai)
Subject: Morphological Subband decomposition
Message-ID: <1995May24.092703.1263@uxmail.ust.hk>
Sender: usenet@uxmail.ust.hk
Nntp-Posting-Host: cssu81.cs.ust.hk
Organization: The Hong Kong University of Science and Technology
X-Newsreader: TIN [version 1.2 PL2]
Date: Wed, 24 May 1995 09:27:03 GMT
Lines: 20

Hello,
 
Currently, I'm doing a project, studying the properties
of an adaptive morphological subband decomposition. I
implemented the MSD proposed by Egger, O., Li, W.,
and Kunt, M. published in IEEE Proceedings Feb., 1995
with title "High Compression Image Coding Using an Adaptive
Morpho....". The performance is very good and all the ringing
effect is gone completely. However, in the paper, it mentioned
that this nonlinear SD is not good for texture image. I have
tried proposed MSD and LSD (Linear SD) on Baboon without
using the adaptive version at all for MSD. So far, I can't
conclude that MSD is not good for texture image. It seems
to me that it outperforms LSD also for texture images.
Any comments? Also, do anyone know the email address of the
author? They should be in Swiss Federal Institute of Technology.
Thanks for reading.
 
William

