Copy Detection of Printed Documents using Camera Phones
In the current advanced and highly changing technological environment, the problem of printed documents’ authentication is often a challenging task. We are still in the era of hardcopy documents and we are very much dependent on them in our daily administrative decisions. Consequently, it is apparent that criminals apply their effort and available resources to create fraudulent documents that look identical to the original for the naked eye. Therefore Digitized Document Frauds (DDF) cost different organisations a lot of money, i.e., Universities, Banks, Insurance companies, hospitals. DDF is defined as the process of manipulation of documents digitally followed by reproduction of the document. To discourage these illegal acts some companies rely on “cost-based deterrents” which unable criminals with limited resources to continue their fraud.
Quisquater et al. (1997) discuss securing a digital camera’s video footages using block hashes. Each frame is divided into non-overlapping blocks of fixed size, and then a digest is generated for each block using a hash function (SHA- Secure Hash Algorithm). The sequence of these digests is embedded into the header of the movie. The only party that knows the secret key is the Trusted Third Party (TTP). Gupta et al (2006) conducted a generic experimental study for investigating digitized printed document frauds. The study was tackled from the forensics perspective. In essence, it mainly exploits the fact that each printer generates a specific noise pattern that is unique to it. As such, the source of the examined printed document can be tracked back. Blanchester (2000) invented a method and a device relating to document authentication via an optical scanner and a database where a device can accept ID cards, extract the face region using a built-in scanner, and communicate that to an internal database to authenticate the card holder. There does not exist any software which works on a mobile-camera for “mobile” instant authentication. This motivates the work of Mobile-phone based authentication that we are proposing here.
This proposal is concerned with a solution tailored to fit into mobile phones and builds on the previous Steganography work of the research team (Cheddad et al, 2009, Cheddad et al, 2010).
Brief Methodology
(1) Carry out a literature review in the areas of steganography, cryptography, mobile devices and image processing.
(2) Embark on requirements analysis to determine the current key factors that allow document recognition
(3) Specify a framework which accommodates the work proposed
(4) Evaluate to ensure the improvements of new framework over existing techniques.
The anticipated outcomes relate to the development of a framework to assist in efficient recognition of fraudulent documents using mobile technology. This mechanism would aim to deliver a highly secure, efficient and low cost software-based solution. The digital watermarks would be recognised by enabled software running on a mobile-phone via image acquisition using the phone’s digital camera.
References
Quisquater, J-J., Macq, B., Joye, M., Degand, N. and Bernard, A. (1997). Practical solution to authentication of images with a secure camera, In I.K. Sethi and R.C. Jain, editor(s), Storage and Retrieval for Image and Video Databases. Volume 3022 , pp: 290-297, SPIE, January 1997.
Gupta, G., Mazumdar, C., Rao, M.S. and Bhosale, R.B., (2006). Paradigm shift in document related frauds: Characteristics identification for development of a non-destructive automated system for printed documents. Digital Investigation, 3 (1) 43-55. Elsevier Science.
Blanchester, T.F. (2000). Method and control device for document authentication. USA Patent number: 6141438, Oct. 31, 2000.
Cheddad, A., Condell, J., Curran, K., Mc Kevitt, P. (2009) A Secure and Improved Self-Embedding Algorithm To Combat Digital Document Forgery. Signal Processing, Special issue on Visual Information Analysis and Security, Vol. 87, No. x, pp: , 10.1016/j.sigpro.2009.02.001, Autumn 2009, ISSN: 0165-1684, Elsevier
Cheddad, A., Condell, J., Curran, K., Mc Kevitt, P. (2010) Towards Objectifying Information Hiding, ICASSP 2010 - The 35th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Sheraton Dallas Hotel, Dallas, Texas, USA, March 14-19 2010
Personnel Involved
First Supervisor: Curran, K Dr
Second Supervisor: Condell, J Dr
Collaboration: This project does not involve collaboration with another establishment