NJIT
Detection and robustness of digital image watermarking signals : a communication theory approach
Abstract
dc:description.abstractThe detection and robustness of the watermark signal is studied from a communications point of view. The contributions of this dissertation are presented in two parts. The first part, which covers the detection aspect, introduces a new digital image watermarking approach that embeds meaningful information in a copyright protection watermark signal; demonstrates the need to approach the watermark signal as a power-constrained signal; studies the relationship between the watermark signal dimension and the image capacity to the signal; explains the similarities and differences between detecting the watermark signal and detecting a signal over a spread-spectrum communication channel; and analyzes the application of sequence detection techniques (MAPSD and MLSD) to the watermark signal. The second part, which covers the robustness aspect, introduces a novel multidimensional interleaving algorithm that increases the signal's robustness against burst errors; presents, analyzes, and compares two techniques for implementing the algorithm (a sliding window technique and a successive partitioning technique); and demonstrates the increase in watermark signal robustness as a result of applying this multidimensional interleaving. This increase of the signal's robustness is shown in the 2-D case by applying the 2-D version of the interleaving algorithm to watermark signals embedded in still images (where the signal layout is in 2-D), and in the 3-D case by applying the 3-D version of the interleaving algorithm to watermark signals embedded in video sequences (where the signal layout is in 3-D).
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy in Electrical Engineering - (Ph.D.)
- Discipline thesis:degree_discipline
- Electrical and Computer Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Elmasry, George F.
- Contributors dc:contributor
-
- Yun Q. Shi
- Michael Fang
- Joseph Frank
Subjects
dc:subject × 4Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.njit.edu/dissertations/970