A Study And Analysis Of Watermarking Algorithms For Medical Images
Loading...
Date
2010-12
Authors
Al-Qershi, Osamah Mohammed Abdo
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
Digital watermarking techniques hide digital data into digital images
imperceptibly for different purposes and applications such as copyright protection,
authentication, and data hiding. One of the objectives of health care management
systems is to securely archive patients’ records. Moreover, these records may require
very large media capacity to store, long time to transmit, and consequently incur
higher cost. Classical encryption technology is an important tool that can be used to
protect data transmitted over computer networks but it does not solve all digital data
protection problems. Digital watermarking is an emerging technology for digital
image authentication. It is proposed to overcome the problems of authentication,
capacity and cost as the watermarking schemes can be used to store different medical
records such as graphs, text, etc. This eliminates storing separate data files along
with the medical images. In this thesis, four digital watermarking algorithms were
proposed for medical images. The first two algorithms are reversible and can be used
for data hiding application. The proposed two algorithms showed high embedding
capacity with very good visual quality. The third algorithm is a reversible algorithm
with authentication, tamper detection, and recovery facilities. The algorithm showed
good performance in terms of hiding capacity and visual quality. Also the proposed
algorithm could authenticate the watermarked images. In case of tampering, the
proposed algorithm could detect the tampered areas and even recover those areas in
some cases. The fourth algorithm is a hybrid algorithm with authentication, tamper
detection, and recovery facilities. The results showed good performance in terms of
the ability to store patient’s data and visual quality. Also the authentication and
tamper detection facilities were tested same as in the third proposed algorithm. In
addition, the algorithm showed some robustness against certain types of noise such
as salt and pepper and cropping. The general performances of the four proposed
algorithms depend on the modality of the image undergone watermarking.
Description
Keywords
A study and analysis of watermarking , algorithms For medical images