Anti-tampering and liveness

What is anti-tampering

Fingerprint recognition is more often being used in mobile and desktop solutions to provide convenient and secure identity authentication. But, today’s fingerprint sensors are vulnerable to spoofing via fake fingers. The risk for fraud and spoofing via fake fingerprints poses a threat to sensitive applications like mobile payments.

A determined and thoughtful person can effectively steal another individual’s fingerprint identity via so called uncooperative spoofing. Fingerprints are left on numerous surfaces and it is possible to readily acquire a latent fingerprint in under 60 seconds using a variety of known methods (e.g., lift tape, cyanoacrylate fuming, photography). Once latent prints are captured and digitized, fingerprint molds and fake fingerprints are readily produced using household materials such as gelatin, latex paint and modelling clay.

What is spoofing

Spoofing is an attack at the sensor level in which a biometric sample is replaced by an imposter’s sample. Susceptibility of biometric scanners to spoof attacks is a well-documented problem.

1. extract information from the material for a template;
2. compare information from templates saved previously;
3. determine whether the biometric information is identical.

Biometric technologies therefore consist of both hardware such as fingerprint sensors or cameras. These can physically read the biometric information. In addition, software is needed that together with the hardware collects the biometric information and then extracts, compares and verifies the information.

Liveness and anti spoof

In addition to verifying that a certain biometric information comes from a certain person, they often also want to ensure that the biometric information comes from a living, physical person. In order to prevent fraudsters from attempting to fool the systems by collecting e.g. a fingerprint from an object and then creating different impressions of e.g. clay, glue, gelatine, etc., different forms of anti-spoof protection are used. These can be directly in the hardware; some sensors can, for example, read whether the material conducts current, like a finger, or illuminate a finger with different lights to see the inner layers of the skin. Other solutions use software optimized to recognize differences in materials. Some systems use a combination of both hardware and software-based anti-spoof protection.