Image compression using neural networks and wavelet transform

Other Title(s)

ضغط الصور باستخدام الشبكات العصبية و التحويل المويجي

Dissertant

Abbas, Ahmad Kamil

Thesis advisor

Jayyad, Adalah Mahdi

University

University of Basrah

Faculty

Science College

Department

Department of Computer Science

University Country

Iraq

Degree

Master

Degree Date

2013

English Abstract

Digital images are widely used in computer applications.

Uncompressed digital images require considerable storage capacity and transmission bandwidth.

Efficient image compression solutions are becoming more critical with the recent growth of intensive data, multimedia-based web applications.

This thesis relies on neural networks as a basis for image compression which covers three categories: image compression using neural networks only, wavelet followed by neural network in which the neural network is used as an encoder to encode the significant coefficients of wavelet transform, and finally neural network followed by wavelet transform.

The third technique is considered as the main idea in this thesis.

In this technique the wavelet is applied on the output of the neural network hidden layer.

This method is trying to exploit the well-known characteristics of wavelet transform in image compression to improve the compression performance obtained by neural network only, while keeping the centrality of neural network in compression system.

Image quality is measured objectively using peak signal-to-noise ratio and subjectively using perception.

The results show the performance superiority of the new method compared to two previously mentioned techniques in the quality of the reconstructed image, computing time, and compression ratio.

The proposed method is implemented for both color image and grayscale images from different classes

Main Subjects

Information Technology and Computer Science

No. of Pages

93

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-743231