Welcome To UTPedia

We would like to introduce you, the new knowledge repository product called UTPedia. The UTP Electronic and Digital Intellectual Asset. It stores digitized version of thesis, dissertation, final year project reports and past year examination questions.

Browse content of UTPedia using Year, Subject, Department and Author and Search for required document using Searching facilities included in UTPedia. UTPedia with full text are accessible for all registered users, whereas only the physical information and metadata can be retrieved by public users. UTPedia collaborating and connecting peoples with university’s intellectual works from anywhere.

Disclaimer - Universiti Teknologi PETRONAS shall not be liable for any loss or damage caused by the usage of any information obtained from this web site.Best viewed using Mozilla Firefox 3 or IE 7 with resolution 1024 x 768.

Image Indexing and Retrieval based on Hybrid Feature Extraction and Classification Using Image Processing and Machine Learning Algorithms

Baharudin, Baharum (2005) Image Indexing and Retrieval based on Hybrid Feature Extraction and Classification Using Image Processing and Machine Learning Algorithms. PhD thesis, University of Bradford.

[img] PDF
Download (3901Kb)


Everyday more images are being created, stored and transmitted. However these three acts themselves do not really pose serious problems. The problem becomes apparent when the stored images need to be retrieved. Query using the traditional text-based approaches, though simple and easy to implement, are no longer sufficient when considering the large volume of images that have to be manually labelled. A logical solution to this problem is to search for images based on its content. Thus ContentBased Image Retrieval (CBIR) was born. Since then, many systems have been developed either commercially or in the form of research prototypes. The heart of any CBIR system is feature extraction. In other words features extracted from the images (usually in the form of a vector representation) becomes the index by which the images will be searched. In terms of the number of features used to represent images, it is generally accepted that the use of multiple image features is more desirable than using a single feature. This is evident judging from the major CBIR systems that have been developed. In this thesis, a hybrid feature extraction scheme is proposed based on a combination of features derived from the compressed as well as the pixel domain. By using two well-known classifiers; the Backpropagation Neural Network and Support Vector Machines, the performance of the proposed hybrid feature approach is compared with that of the other feature based approaches which serve as benchmarks. From the results obtained it has been shown that the hybrid feature extraction approach outperforms all the other feature based methods used in the experiments.

Item Type: Thesis (PhD)
Academic Subject : Academic Department - Electrical And Electronics - Instrumentation and Control - Intelligent System - Imaging in intelligent surveillance systems
Subject: T Technology > TJ Mechanical engineering and machinery
Divisions: Engineering > Electrical and Electronic
Depositing User: Users 2053 not found.
Date Deposited: 27 Sep 2013 11:00
Last Modified: 25 Jan 2017 09:46
URI: http://utpedia.utp.edu.my/id/eprint/6925

Actions (login required)

View Item View Item

Document Downloads

More statistics for this item...