Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques

The topic of image compression and retrieval has become one of the most researched areas in the recent years due to the acute demand for storage and transmission of large volume of image data that are generated in the Internet and other applications. When compressing an image, it is necessary to sat...

Full description

Saved in:

Bibliographic Details
Main Author:	Ahamed Ayoobkhan, Mohamed Uvaze
Format:	Thesis
Published:	2017
Subjects:	TA1501-1820 Applied optics Photonics
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-mmu-ep.7154
record_format	uketd_dc
spelling	my-mmu-ep.71542024-08-19T03:41:57Z Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques 2017-06 Ahamed Ayoobkhan, Mohamed Uvaze TA1501-1820 Applied optics. Photonics The topic of image compression and retrieval has become one of the most researched areas in the recent years due to the acute demand for storage and transmission of large volume of image data that are generated in the Internet and other applications. When compressing an image, it is necessary to satisfy two conflicting requirements, namely, compression ratio (CR) and the image quality which is usually measured by the parameter, peak signal-to- noise ratio (PSNR). In this thesis, several lossless and lossy image compression techniques as well as an integrated image retrieval system are proposed using prediction and wavelet based techniques. Employing prediction errors instead of the actual image pixels for compression and retrieval processes ensures data security. A lossless algorithm (LLA) is proposed which uses neural network predictors and entropy encoding. Classification is performed as a pre-processing step to improve the compression ratio. For this purpose, classification algorithm1(CL1) and classification algorithm2(CL2) which make use of wavelet based contourlet transform coefficients and Fourier descriptors as features are proposed. Two identical artificial neural networks (ANNs) are employed at the compression (sending) and decompression (receiving) sides to carry out the prediction. The prediction error which is the difference between the original and the predicted pixel values is used instead of the actual image pixels. The prediction is performed in a lossless manner by rounding-off the predicted values to the nearest integer values at both sides. 2017-06 Thesis https://shdl.mmu.edu.my/7154/ http://erep.mmu.edu.my/ phd doctoral Multimedia University Faculty of Computing and Informatics EREP ID: 1740
institution	Multimedia University
collection	MMU Institutional Repository
topic	TA1501-1820 Applied optics Photonics
spellingShingle	TA1501-1820 Applied optics Photonics Ahamed Ayoobkhan, Mohamed Uvaze Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
description	The topic of image compression and retrieval has become one of the most researched areas in the recent years due to the acute demand for storage and transmission of large volume of image data that are generated in the Internet and other applications. When compressing an image, it is necessary to satisfy two conflicting requirements, namely, compression ratio (CR) and the image quality which is usually measured by the parameter, peak signal-to- noise ratio (PSNR). In this thesis, several lossless and lossy image compression techniques as well as an integrated image retrieval system are proposed using prediction and wavelet based techniques. Employing prediction errors instead of the actual image pixels for compression and retrieval processes ensures data security. A lossless algorithm (LLA) is proposed which uses neural network predictors and entropy encoding. Classification is performed as a pre-processing step to improve the compression ratio. For this purpose, classification algorithm1(CL1) and classification algorithm2(CL2) which make use of wavelet based contourlet transform coefficients and Fourier descriptors as features are proposed. Two identical artificial neural networks (ANNs) are employed at the compression (sending) and decompression (receiving) sides to carry out the prediction. The prediction error which is the difference between the original and the predicted pixel values is used instead of the actual image pixels. The prediction is performed in a lossless manner by rounding-off the predicted values to the nearest integer values at both sides.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Ahamed Ayoobkhan, Mohamed Uvaze
author_facet	Ahamed Ayoobkhan, Mohamed Uvaze
author_sort	Ahamed Ayoobkhan, Mohamed Uvaze
title	Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
title_short	Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
title_full	Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
title_fullStr	Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
title_full_unstemmed	Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques
title_sort	image compression and retrieval using prediction and wavelet - based techniques
granting_institution	Multimedia University
granting_department	Faculty of Computing and Informatics
publishDate	2017
_version_	1811768021710012416

Image Compression And Retrieval Using Prediction And Wavelet - Based Techniques

Similar Items