AAU-ETD AAU-ETD
 

Addis Ababa University Libraries Electronic Thesis and Dissertations: AAU-ETD! >
Faculty of Informatics >
Thesis - Computer Science >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/1321

Title: Shape Sensitive Salient-object based Image Query
Authors: Estifanos, Gashaw
Advisors: Dr. Solomon Atnafu
Keywords: Salient-Object Based Image Retrieval
Image Database
Image Retrieval in Pervasive Environment
Image Segmentation
Multi-Criteria Image Query
Copyright: 2005
Date Added: 29-Aug-2008
Publisher: Addis Ababa University
Abstract: The availability of images in electronic format has resulted in large number of digital images being used for different applications. The growth in the usage and availability of digital images calls for an efficient means of image data management. This has led to the development of content based image retrieval (CBIR) systems, which uses visual contents of an image such as color, texture or shape for querying image databases. CBIR is well researched and there are a large number of image retrieval systems based on this technology. Image retrieval matching can be performed using global features of the whole image (as in CBIR), but a comparison that considers part(s) of images for similarity-based retrieval is sometimes a more natural approach. Regions of an image that are of particular interest to humans are termed salient-objects of an image. Salient –object based image retrieval is not well researched, but there some developed system that model the salient object of the image using a Minimum Bounding Rectangle. In this work a new approach is proposed which considers the natural shape of the salient objects contained within an image. An Image segmentation tool is used to automatically identify salient objects of an image and store them in an object relational DBMS for multi-criteria image query (query based on both image content, positional predicates and metadata). Furthermore, some techniques have been proposed in this work to make the image retrieval system accessible within a pervasive computing environment.
Description: A Thesis Submitted to the School of Graduate Studies of Addis Ababa University in partial fulfillment of the requirements for the Degree of Master of Science in Computer Science
URI: http://hdl.handle.net/123456789/1321
Appears in:Thesis - Computer Science

Files in This Item:

File Description SizeFormat
Estifanos.pdf2.15 MBAdobe PDFView/Open

Items in the AAUL Digital Library are protected by copyright, with all rights reserved, unless otherwise indicated.

 

  Last updated: May 2010. Copyright © Addis Ababa University Libraries - Feedback