A new touchless, three-dimensional fingerprinting system promises faster and more accurate processing than the ink-and-roll solution used by many law enforcement agencies today.
The new system, developed by University of Kentucky researchers, projects patterns of light on a finger and analyzes the resulting image. They say the system is more efficient and significantly reduces the number of incorrect matches.
The researchers have formed a company, FlashScan3D, to market the product. According to Ph.D student and paper co-author Yongchang Wang, the scanner has been requested by the Smithsonian for its collection.
Wang says the current scanner is “portable,” and takes seven-tenths of a second to scan one 3D fingerprint with 5 million pixels. The 3D fingerprints are so detailed that even pores on fingers are clear, he says.
The latest approach to taking fingerprints involves rolling a subject’s finger over a glass plate for scanning, but that method can require several attempts per finger to get a usable print. The whole process can take several minutes.
Technology Review reports:
“The customs agent has a budget of 32 seconds per person. They need a way to get your fingerprints quickly,” says Mike Troy, chief executive officer of FlashScan3D, a company based in Richardson, TX, that was founded to commercialize the Kentucky system.
Specifically, the scanner works by projecting a series of striped lines onto a finger, called “structured light illumination.” A 1.4 megapixel camera captures images of the lines as they wrap around the finger, at roughly 1,000 pixels per inch — twice the minimum resolution required by the FBI.
Guided by the movement of the light over the finger, the software builds a three-dimensional model of the ridges and valleys of the finger’s surface.
Better still, this system has none of the print distortion associated with forensic methods that require rolling a finger on a surface.
The Department of Homeland Security and the National Institutes of Justice have expressed interest in a non-contact system that can capture 3-D prints, ideally gathering data from multiple fingers at once.
The interest — and grant money attached to it — has sparked more research into the area: for example, Carnegie Mellon University and TBS Holdings are independently working on systems that use multiple cameras to capture the prints.
The challenge for everyone? To turn that three-dimensional data into 2-D prints to work with existing technology.
