Use this URL to cite or link to this record in EThOS:
Title: Watermarking and Steganography of 3-D models
Author: Salman, Muhammad
ISNI:       0000 0004 2684 1529
Awarding Body: University of Surrey
Current Institution: University of Surrey
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Access from Institution:
Creation of 3-D digital models is an expensive process and the owners of these models are reluctant to make the models publically available. A typical example is the case of interactive web applications where these models can be downloaded and used illegally without their consent. Watermarking addresses the issue of copyright protection of 3-D models by inserting the copyright information (watermark) robustly and invisibly in 3-D models without creating a visually discernable impact on the model. This watermark may be useful as a proof of ownership of a 3-D model. The first contribution of this thesis is the design and development of a novel watermarking system for copyright protection of 3-D triangle models. The proposed method inserts the watermark in the normal vector information of 3-D triangle models, which is least likely to be disturbed by the attacks due to its importance in the rendering process. Experiments prove that proposed method is robust against the cropping, mesh simplification and noising attacks in addition to rotation, translation and uniform scaling. The remaining part of the thesis addresses issues regarding the steganography of 3-D models. Steganography is described as hiding the secret message in 3-D model in a manner that the presence of the secret message in 3-D model is undetectable. An efficient steganographic method is proposed for 3-D polygonal and point sampled models which exploits the Implicit Point Order (IPO) of 3-D models to compute the fictitious list of the vertices of the model so as to embed the hidden message in 3-D model. Experiments show 30% improvement in the efficiency as compared to the state of the art techniques. Another high capacity steganographic algorithm is designed and implemented for 3-D point sampled and polygonal models which embeds the secret message in the least significant bits of the transformed coordinates of the point data. Capacity has been improved to 30 bits per vertex at some expense of the complexity. However, the processing time is still within the practical limits.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available