Image processing beats paper forgery

To counter the problem of document forgery, researchers at New York University have developed a tamper-resistant paper watermarking technique that extracts a unique watermark from any piece of paper based on the natural randomness present in the structure of the paper.

Forgery of paper documents is prevalent in many countries around the world, especially in developing regions.

To counter the problem, researchers at New York University (New York, NY, USA) have developed a tamper-resistant paper watermarking technique that extracts a unique watermark from any piece of paper based on the natural randomness present in the structure of the paper.

Unlike earlier techniques that extracted watermarks based on the fiber structure of paper, the so-called PaperSpeckle method captures a random speckle pattern -- a random bright/dark region formation at the microscopic level when light falls on to the paper -- and then uses image-processing algorithms to generate a unique fingerprint from it.

They believe that the technique can be used to make any piece of paper self-verifiable by first extracting the speckle pattern from a region on the paper and then imprinting the fingerprint of the pattern on the same paper using a QR (Quick Response) code.

The researchers have demonstrated the PaperSpeckle technique on a laptop PC attached with a USB microscope and on a Google Nexus One mobile phone with a microscope attached to its camera.

They have tested it on different types of paper and have shown that it can produce a robust repeatable watermark even if paper is damaged due to folding, crumpling, soaking in water/oil, or ageing with time.

The researchers will discuss the system in detail at the Computer and Communication Security conference in Chicago later this month.

A detailed description of the PaperSpeckle technique is available.

-- By Dave Wilson, Senior Editor, Vision Systems Design

More in Security, Surveillance, Transportation