BiometricsResearchers develop altered fingerprint detector

Researchers at Michigan State University have developed a technique to help detect when an individual has deliberately altered their fingerprints in an effort to fool biometric scanners.

With biometric passports and fingerprint scanners on the rise around the world, more and more individuals have turned to extreme measures to avoid detection like removing the pads of their fingertips or surgically replacing fingerprints with toe prints.

For instance, in 2009, a Chinese woman illegally entered Japan by altering her fingerprints to fool immigration officials.

To help law enforcement and immigration officials combat this growing trend, researchers have developed an algorithm to help identify altered prints.

According to Soweon Yoon, a graduate student working with Anil Jain, a biometrics expert at Michigan State University who is heading up the project, spotting altered fingerprints is not easy, but recognizing when there are an unusual amount of anomalies is a possibility.

“The most important feature for fingerprint matching is called minutiae. So minutiae refers to [a] ridge ending point and [a] ridge bifurcation point,” Yoon explained. “From typical fingerprint impression, we can extract 100 minutiae per each finger.”

Unlike normal prints, altered fingers have abrupt, discontinued lines which usually contain a high number of minutiae points. Individuals will often cut a “Z” shape into their hands, burn it with acid, or even swap with another person.
“You can imagine if someone makes a cut here, they will generate a lot of ridge ending points, so that generates [an] excessive number of minutiae,” Yoon said.

With existing software, which searches through millions of fingerprints, spotting anomalies can be nearly impossible, Yoon said.

“Some of these databases are extremely, extremely large. So the FBI database is an example, has about 70 million subjects. That means there are 700 million fingerprint images if each person has 10 fingers. And so there is no way anybody could manually check whether there are altered fingerprints or not,’ Yoon said.

Rather than identifying people, the altered fingerprint algorithm focuses on alerting officials if there are any anomalies so they can perform a second inspection.

The algorithm is also careful to distinguish between accidental alterations and intentional ones as accidents result in scars on only a few fingers, so if there are changes on five or more fingers, the alteration was likely done on purpose.

Jain and her team have licensed the new technique to Morpho, a major French biometric solutions developer. The company is confident that it can easily attract interested buyers in the new application as it currently serves many of the world’s major law enforcement agencies.