A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector

Recent advances in technology have enabled the digitization of medical images. However, due to the confidentiality requirement for medical images there is the need to ensure that digital medical images are protected from illegal distortions and reproduction. In this research, a new watermarking tech...

Full description

Saved in:
Bibliographic Details
Main Author: Hajiabdollah, Tina
Format: Thesis
Language:English
Published: 2012
Subjects:
Online Access:http://eprints.utm.my/id/eprint/41581/1/TinaHajiabdollahMFSKSM2012.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my-utm-ep.41581
record_format uketd_dc
spelling my-utm-ep.415812017-09-20T05:43:23Z A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector 2012-12 Hajiabdollah, Tina QA76 Computer software Recent advances in technology have enabled the digitization of medical images. However, due to the confidentiality requirement for medical images there is the need to ensure that digital medical images are protected from illegal distortions and reproduction. In this research, a new watermarking technique based on spatial domain for digital mammogram images is proposed. In the proposed technique, the ROI and RONI of the mammographic image are detected using the Sobel edge detector. Afterwards, the image is partitioned into 16 blocks of the same size, where percentage value of white pixels is calculated for each block. The blocks with the lowest percentage are chosen, which in this case two blocks appear to be of lowest percentage values. The two blocks are of two percent (2%) values each and are used for the embedding process. We then apply BiISB that makes use of two levels of ISB bit-planes (which are the higher-order-ISB and low-order-ISB) at the embedding and extracting phases. The BiISB is employed so as to ensure that the watermarking process is robust and the digital images are made to be imperceptible. For ensuring the security of the digital images, three secret keys were employed at the extraction phase. To verify the robustness and transparency of the proposed technique, NCC and PSNR values are calculated for some sampled images. The results show that the NCC values for the proposed technique maintained high robustness for median filtered images, Salt and Pepper images, Speckle images, Gaussian images, motion effect images, Contrast enhanced images, whereas, the Poisson noise images, rotation and JPEG compressed images maintained values between 0.76 and 1.0. By averaging the PSNR values of all the sampled images, a PSNR value of 59.26 db was achieved. 2012-12 Thesis http://eprints.utm.my/id/eprint/41581/ http://eprints.utm.my/id/eprint/41581/1/TinaHajiabdollahMFSKSM2012.pdf application/pdf en public masters Universiti Teknologi Malaysia, Faculty of Computing Faculty of Computing
institution Universiti Teknologi Malaysia
collection UTM Institutional Repository
language English
topic QA76 Computer software
spellingShingle QA76 Computer software
Hajiabdollah, Tina
A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
description Recent advances in technology have enabled the digitization of medical images. However, due to the confidentiality requirement for medical images there is the need to ensure that digital medical images are protected from illegal distortions and reproduction. In this research, a new watermarking technique based on spatial domain for digital mammogram images is proposed. In the proposed technique, the ROI and RONI of the mammographic image are detected using the Sobel edge detector. Afterwards, the image is partitioned into 16 blocks of the same size, where percentage value of white pixels is calculated for each block. The blocks with the lowest percentage are chosen, which in this case two blocks appear to be of lowest percentage values. The two blocks are of two percent (2%) values each and are used for the embedding process. We then apply BiISB that makes use of two levels of ISB bit-planes (which are the higher-order-ISB and low-order-ISB) at the embedding and extracting phases. The BiISB is employed so as to ensure that the watermarking process is robust and the digital images are made to be imperceptible. For ensuring the security of the digital images, three secret keys were employed at the extraction phase. To verify the robustness and transparency of the proposed technique, NCC and PSNR values are calculated for some sampled images. The results show that the NCC values for the proposed technique maintained high robustness for median filtered images, Salt and Pepper images, Speckle images, Gaussian images, motion effect images, Contrast enhanced images, whereas, the Poisson noise images, rotation and JPEG compressed images maintained values between 0.76 and 1.0. By averaging the PSNR values of all the sampled images, a PSNR value of 59.26 db was achieved.
format Thesis
qualification_level Master's degree
author Hajiabdollah, Tina
author_facet Hajiabdollah, Tina
author_sort Hajiabdollah, Tina
title A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
title_short A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
title_full A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
title_fullStr A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
title_full_unstemmed A new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
title_sort new mammogram image watermarking technique based on two-level intermediate significant bits and sobel edge detector
granting_institution Universiti Teknologi Malaysia, Faculty of Computing
granting_department Faculty of Computing
publishDate 2012
url http://eprints.utm.my/id/eprint/41581/1/TinaHajiabdollahMFSKSM2012.pdf
_version_ 1747816575549833216