Beware evil villains and career crooks, the machine vision technology company BitFlow just developed a revolutionary fingerprint scanner capable of discerning fake fingerprints from real ones. Whereas typical fingerprint scanners are fairly simple to fool, BitFlow's new tech scraps the traditional process of photo scanning in favor of a method which actually looks inside a person's finger. Time to throw out all that Play-Doh you had reserved for shady fingerprint forgery.
Developed in conjunction with scientists at the Langevin Institute in Paris, France, the innovative new scanner quickly and accurately captures, what the team calls, a person's "internal fingerprint." Similar to fingerprints seen on the surface of a finger, an internal fingerprint has the exact same topographical features, yet reside roughly a half millimeter below the skin. Moreover, the system also boasts the capability to image sweat pores in a person's finger, further enhancing its means of identifying someone.
Engineered by Langevin Institute postdoctoral researcher Egidijus Auksorius and scientific instrument professor Claude Boccara, the internal sensor is based on full-field optical coherence tomography -- or FF-OCT. Unlike standard optical coherence tomography, which makes use of 3D data and lasers, FF-OCT relies on a 2D detector, making it easier and faster to use (not to mention, cheaper). Basically, images of an internal fingerprint can be captured in under a second using this tech which is an absolute boon for the fingerprint scanning industry.
When in use, the scanner analyzes the various interference patterns produced after a beam of light reflects off a sample (i.e. a finger) against a reference beam of light. For instance, once a light beamed at a finger registers with the system, it has the ability to see the different grooves, patterns, and pores associated with someone's unique fingerprint. The reference beam of light -- which is just a pure beam -- allows the scientists to accurately read the differences and chart the grooves and patterns.
According to BitFlow's published press release, a study conducted by the Department of Homeland Security found wearing fake prints, or "spoofing" a fingerprint sensor, beats a scanner roughly 80 percent of the time. Considering this is a staggering percentage in any capacity, Boccara and Auksorius' collaboration with BitFlow appears poised to be a forensics expert's best friend.