Color models indexing for content based image retrieval system

Dissertant

Ibrahim, Ibrahim Nadhir

Thesis advisor

Salih, Hilal Hadi

University

University of Technology

Faculty

-

Department

Computer Sciences Department

University Country

Iraq

Degree

Ph.D.

Degree Date

2009

English Abstract

In this thesis a Content-Based Image Retrieval {CB1R.} system is presented that supports querying with respect to color and texture low-level features.

The fundamental idea is to generate automatically image descriptors by analyzing the image content.

The underlying techniques are hase on the adoption oTquantization techniques of HSV, YIQ, and RGB color models with the Gray-Level Run Length Matrix (GL&Lhf), these techniques are use=d to extract sets of (texture and color) features.

These fea.lu.res are used for retrieval tasks in separated and combined manners.

The clustering quality criteria concepts have been utilized to reduce the number of possible feature combinations, Each image is represented by features vector(s) in the features space.

These vectors are indexed using a partitioned clustering algorithm called K-mean Clustering which provides casy-to-index data structures as well as faster query execution facilities.

The degree of similarity between images is defined by the distance in the features space.

Given a query image, the system first extracts its features vector, and then compares this vector with those of the images pointed along the index structure using Euclidean distance measure.

In this way, the matched images could be ranked and put into group according to the distance -of their features vectors to the query one.

This ranked ^roup i-s considered as the query result.

The performance of retrieval system has been evaluated using two measurements (i.e., precision and recall).

During the evaluation process a comparison study is made between different applied retrieval schemes.

The HSV color representation wilh run-length features is the best from: speed and accuracy point of view -when using single (around 0.63 for precision..

and 0.12 for recall) ot combined two types of features (around 0.67 hbr precision, and 0.13 for recall) or three types of features (around 0.09 for precision, and 0.14 for recall).

But in this situation while the accuracy of system has been increased the time needed for system execution will also increased

Main Subjects

Information Technology and Computer Science

Topics

• Data processing
• Digital electronics

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-305266