Principal Investigator: Edward Delp
In today's digital world securing different forms of content is very important in terms of protecting copyright and verifying authenticity. The ability to identify the device or type of device used to print material in question would provide a valuable aid for law enforcement and intelligence agencies. We have developed two strategies for printer identification based on examining a printed document. The first involves characterizing the printer by finding intrinsic features in the printed document that are characteristics of that particular printer, model, or manufacturer's products. Using this "intrinsic signature" allows up to 90% classification accuracy among documents printed from multiple printers of different ages, and on different paper types. The second strategy embeds an "extrinsic signature" in a printed page by modulating the process parameters in the printer mechanism to encode identifying information such as printer serial number and date of printing. Performing the embedding in the printer allows much finer control over the marks that can be placed on the page and also increases the difficulty of "hacking" the system. Using extrinsic signatures allows embedding of up to 400 bits in a page of text. Further analysis of the extrinsic embedding parameters we believe will allow us to increase robustness of this embedding scheme and allow increased embedding rates.
Other PIs: Jan P. Allebach (Purdue University) George Chiu (Purdue University)
A. K. Mikkilineni, N. Khanna and E. J. Delp, " Texture Based Attacks on Intrinsic Signature based Printer Identification", Proceedings of the SPIE Media Forensics and Security II, San Jose, CA, January 2010.
Keywords: counterfeiting, device, forensics, forgery, signatures