Content-based image retrieval system

Pao Yue-kong Library Electronic Theses Database

Content-based image retrieval system

 

Author: Lee, Chung-tak
Title: Content-based image retrieval system
Degree: M.Sc.
Year: 1999
Subject: Image processing
Pattern recognition systems
Hong Kong Polytechnic University -- Dissertations
Department: Multi-disciplinary Studies
Dept. of Electronic Engineering
Pages: v, 50 leaves : ill. ; 31 cm
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1479260
URI: http://theses.lib.polyu.edu.hk/handle/200/3113
Abstract: As more and more image data are being generated, managing and manipulating them becomes an important issue to be resolved before we can take full advantage of their information content. A technique called Content-based Image Retrieval (CBIR) therefore needs to be evolved. CBIR System is playing an important role in managing and manipulating image information. This is an alternative approach to text-based system and working with descriptions based on properties which are inherent in the images themselves. The idea behind this is that the natural way to retrieve visual data is by a query based on the visual content of an image: the patterns, colours, textures, and shapes of image objects, and related layout and location information. The queries of this type may be either supplemental or preferable, to text, and in some cases they may be necessary. Furthermore, visual queries may simply be easier to formulate. In general, the issues addressed by this kind of system which supports queries based on image content may be summarized as follows: - Selection, derivation, and computation of image features and objects that provide useful query expressiveness; - Retrieval methods based on similarity, as opposed to exact matching. For this project, 4 schemes for constructing a feature/image database for query processing were studied. Database query is performed by feeding an image as input to the system. Different features are extracted from the image and then used for matching pictures of similar features in the database. The feature that we are focusing on is the image texture property. Four schemes which include - Mean of pixel values. - Standard deviation of pixel values. - Histogram of the whole image. - 2-D circularly symmetric Gabor filter. have be implemented. These features are to be compared with a given set of images for the accuracy representing the image - rate of image/pattern recognition. The detail implementation is shown in chapter 3 and the performance of each scheme has been investigated and presented in the chapter 4.

Files in this item

Files Size Format
b14792606.pdf 1.914Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information