Please use this identifier to cite or link to this item:
Title: Watermarking techniques using knowledge of host database
Authors: SUJOY ROY
Keywords: Copy Detection, Nearest Neighbor Search, Resolving Ambiguity, Retrieval, Watermarking, Database, Non-convex optimization.
Issue Date: 19-Jul-2006
Citation: SUJOY ROY (2006-07-19). Watermarking techniques using knowledge of host database. ScholarBank@NUS Repository.
Abstract: Many watermarking applications deal with a database of hosts. When given a database of hosts to be watermarked, under the traditional watermarking approach, every host is independently watermarked. That is, encoding of one host does not use the knowledge of existence of other hosts in the database. If the encoder knows in advance about all the hosts in the database to be watermarked, intuitively, it has more information and hence can perform better. However it is not clear how to exploit this information and how significant is the improvement. In this thesis, we propose the notion of knowledge of hosts database and address this question: ``If the encoder has prior knowledge of the hosts database, and the detector has full or partial information of the hosts database, how to exploit this additional information to significantly enhance performance''. To handle this question, a novel approach that demonstrates the efficacy of using knowledge of hosts database during the watermarking process is proposed. The proposed approach is generic and based on this, frameworks that address the problems associated with different applications can be designed. In this dissertation three different frameworks are proposed for three different applications, namely, copy detection, retrieval systems and database watermarking. In each case, novel methods are designed to implement each framework. Systematic theoretical formulation and practical experimental evaluation is performed to validate the efficacy of the proposed frameworks.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
dissertation.pdf3.23 MBAdobe PDF



Page view(s)

checked on Apr 8, 2019


checked on Apr 8, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.