الفهرس 
INTRODUCTION TO DIGITAL IMAGE WATERMARKINGOnly 14 pages are availabe for public view
 |  | from | 94 |  |  |
| | | from | 94 | | |
Abstract
Since multimedia data is now more widely distributed in digital form with the advances in mobile and internet technologies and it is easier to copy and change, copyright protection has become an essential need. There are several techniques to safeguard intellectual property. Such techniques encompass Digital Rights Management, which utilizes encryption and licensing to deter unauthorized content usage; Content Fingerprinting, which generates a distinct signature for tracking content dissemination; and Copy Detection Mechanisms, aimed at thwarting unauthorized content replication. Among these, Digital Watermarking has emerged as the predominant method, gaining prominence for its efficacy. It entails the embedding of a unique identifier within media, facilitating authenticity verification and making it the most adopted technique among contemporary copyright protection strategies. This identifier is known as a watermark, which is embedded into another digital signal such as an image, audio, or video in such a way that it can be extracted later for verification or authentication purposes. The embedded information is not significant by itself, but it provides insight into digital media, so it can be used to identify the signal source or verify that the signal has not been tampered with. Watermarks can be used to protect copyrights, track ownership, or prevent unauthorized use of digital data. Invisibility and robustness are crucial aspects of watermarking, and the trade-off between them is challenging. Invisibility refers to the ability of the watermark to be imperceptible to the human eye. While robustness refers to the ability of the watermark to withstand attacks. Attacks include image compression, cropping, noise addition, geometric, and filtering attacks. Despite the wealth of image watermarking methods available in the current literature, several gaps and disadvantages remain prevalent in the domain. Among these, the most significant include the need to trade-off between robustness and invisibility, inefficiencies in computational demand, and vulnerabilities to multifaceted attacks. The current thesis introduces an innovative image watermarking approach, especially tailored to address and bridge these gaps. This thesis presents a new image watermarking method that seeks to break the traditional compromise between robustness and invisibility, aiming to enhance both metrics simultaneously. By using the Schur decomposition and integrating a dynamic weighting factors matrix in the embedding phase following the execution of multiple levels of discrete wavelet transform (DWT) and singular value decomposition (SVD), the algorithm exhibits increased resistance to a variety of attacks. Notably, the innovative embedding formula involves two scaling parameters, a modification that augments imperceptibility without negatively affecting robustness, even with medical images. Furthermore, trajectory-based optimization allows for rapid determination of the optimal values of these scaling factors, thus reducing computational demands. This contrasts with the more prevalent population-based algorithms that are commonly used to determine the optimal scaling factors having high computational costs. To ascertain the robustness and efficiency of our approach, numerous experiments are conducted. The system is tested under varying conditions and attack scenarios with different sizes and shapes of the watermark. The results, which are captured in diverse tables and figures, conclusively showcase the algorithm’s superior performance. Statistical assessments, particularly the Friedman test, further validate the superiority of the proposed method over existing watermarking techniques in terms of invisibility and robustness. The thesis is concluded with prospects for future advancements, such as broadening the watermarking approach to accommodate audio and video content, and the potential for integrating enhanced encryption as well as embedding multiple watermarks.