This dissertation covers research results in three fields, i.e. reverse engineering digital watermarking systems, security in steganography, and camera related digital image processing.
The research in the first field is under a new attacking philosophy, i.e. learn the target watermarking system first before attacking. A 3-stage model is introduced to model generic watermark detection systems so as to aid reversing in a systematic manner. Two reversing techniques, i.e. super-robustness and noise snakes, are developed and applied to real life attacks. In order to handle watermark residing in feature space of very large dimensionality, a more precise geometric model for normalized correlation based watermark detector is presented and analyzed.
The research in the second field is under the philosophy of creating a stego-friendly cover which would ease the embedding restriction. As a theoretic trial, the subset selection technique is examined under the i.i.d. and the first order Markov case. As a practical implementation, this philosophy is implemented in the Apple iChat video-conferencing software via cultural engineering. The goal of this cultural engineering approach is to render covert data to be culturally undetectable instead of statistically undetectable.
The research in the last category includes two topics, i.e. DSLR lens identification and benchmarking camera image stabilization systems. In both applications, white noise pattern image is used as shooting target to supersede the commonly used checkerboard pattern image, since it excels in position registration and bears the histogram-invariant feature in each scenario respectively. For that task of distinguishing different copies of the same lens using chromatic aberration, focal distance is determined to be another important factor in shaping the lens chromatic aberration pattern, which completes the lens identification feature space. For the task of evaluating camera IS systems, a camera-motion blur metric directly generated from test picture histogram is developed to fulfill the benchmarking task without resorting to either human visual assessment or other more complex motion blur metrics.
|Advisor:||Craver, Scott A.|
|Commitee:||Chiu, Kenneth, Fowler, Mark, Fridrich, Jessica|
|School:||State University of New York at Binghamton|
|School Location:||United States -- New York|
|Source:||DAI-B 72/12, Dissertation Abstracts International|
|Keywords:||Cultural engineering, Digital watermarking, Lens identification, Reverse engineering, Steganography, White noise|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be