printer friendly version

True identity put to test

Technology related to physical security systems has made great leaps forward, yet traditional authentication methods such as locks and keys, PINs, and cards are still widely used. These methods increase the risk of credential breach where credentials are passed to others or are lost or stolen, ultimately leading to a compromise in security.

Biometric authentication can easily resolve such problems, drastically enhancing security and the credibility of personal authentication by forcing the use of true identity and preventing people from authenticating themselves using someone else’s credentials.

The advantages provided by biometric technology have allowed for quick adoption by security markets. Among the biometric technologies, fingerprint technology offers the most flexibility with cost benefits that enable its adoption into various applications. Today, fingerprint is the most flexible and reliable method with higher recognition rates than other biometric technologies such as iris, facial, and vein recognition.

However, there is a growing concern with regards to the use of fake fingerprints, since fingerprint residue can easily be captured off of things that we touch in our daily lives. If we do not pay enough attention, somebody can capture and replicate your fingerprints and use them for malicious purposes.

Fake fingerprints and spoofing

Fake fingerprints are made from materials like clay, gelatine, silicone and rubber. Authenticating with these fake fingerprints is called ‘spoofing’. After the release of the iPhone 5S and the introduction of a built-in fingerprint sensor, contests to crack the device’s fingerprint scanner have been held with many hackers joining the competitions and several IT magazines reporting on the vulnerability of fingerprint sensors against fake fingerprints. Numerous videos have recently been uploaded onto YouTube about hackers that have breached the security by making fake fingerprints from Play-Doh, gelatine, silicone, rubber and the likes – including explanations on how to make these fake fingerprints.

What is the solution?

Suprema’s Live Finger Detection (LFD) technology is based on the analysis of the dynamic and static image characteristics of fake fingers and how they can be distinguished from those of live fingers. Using an advanced analysis algorithm, abnormalities are detected in the dynamic changing patterns of fingerprint images and several static features of liveness or unnaturalness of fingers, clearly distinguishing fake fingers from live fingers.

Dynamic changing pattern analysis

As fingers gradually make contact with the sensor surface, live fingers naturally demonstrate changes in patterns of area, intensity, and movement, but fake fingers produce unnatural changing patterns of separated areas, partially dark areas, distorted boundary shape, and a large movement in the core part. By detecting these abnormalities in dynamic changing patterns from continuous analysis of fingerprint images, fake fingers are distinguished from live fingers. Specifically, this method is very effective in rejecting fake fingers made from hard materials such as paper, film, clay, and hard rubber.

Liveness feature analysis

In fingerprint images, there are several localised features which reveal the liveness of fingers: pore distribution, ridge sharpness, regularity of ridge-valley boundary, amongst others. These localised liveness features are normally too small and elaborate to be copied by simple and soft faking materials such as silicon, rubber and gelatine. Since Suprema’s high performance imaging sensor can capture high quality fingerprint images, various local liveness features are easily distinguished by our advanced analysis algorithm.

Unnaturalness feature analysis

Usually, it is very hard to make a perfect fake finger and almost every fake finger cannot avoid revealing its unnaturalness – unnatural sharp boundaries, too many white blobs or too large black blobs within a fingerprint area, abnormal peaks in histogram distribution and so on. By observing the mixture of various unnaturalness features, numerous fake fingers are effectively rejected.

Liveness Decision Engine

Suprema’s newly developed Liveness Decision Engine (LDE) can effectively prevent spoofing. It detects fake fingerprints with a technology called Dual Light Source Imaging which utilises infrared rays and a white light.

The LDE can block fake fingerprints made from paper, film, glue, rubber, clay and silicon by comparing images obtained with white lights and infrared rays.

The new OP5 sensor applied to Suprema’s recently released fingerprint readers has reduced distortion and improved contrast uniformity, and features an Adaptive Gain Control algorithm and a proximity sensor, which enables the sensor to detect fake fingerprints made from paper, film, glue, rubber, clay and silicon all together.

Share this article:

Further reading: