From nobody Fri Apr  4 15:09 EST 1997
Date: Fri, 4 Apr 1997 15:09:31 -0500 (EST)
From: uid no body <nobody>
To: techreps@cs.buffalo.edu
Subject: techrep: POST request
Content-Type: text
Content-Length: 1326

ContactPerson: azhang@cs.buffalo.edu
Remote host: mekab.cs.buffalo.edu
Remote ident: gsesfah
### Begin Citation ### Do not delete this line ###
%R 97-04
%U /projects4/gsesfah/papers/export/VLDB-97/paper.ps
%A Sheikholeslami, Gholamhosein
%A Zhang, Aidong
%T A Clustering Approach for Large Visual Databases
%D February 21, 1997
%I Department of Computer Science, SUNY Buffalo
%K Image clustering
%X The representation and organization of images in a large-scale 
image database is crucial to support effective and efficient 
accesses to image data on the basis of content. 
In this process, significant features must first be extracted from 
image data in their pixel format. These features must then be classified 
and indexed to assist efficient retrieval of image content. 
In this paper, we investigate effective image data representation 
and organization approaches for large-scale image databases.  
An effective block-oriented image decomposition structure is used 
as a fundamental data model for representing image content. 
A clustering mechanism is then proposed to categorize images based on  
feature similarity. Efficient image retrieval is supported. 
Experimental analysis are conducted and presented to demonstrate 
the effectiveness and efficiency of the approaches.