Contactless fingerprinting maturation allows the unification of biometric capture using smartphones
Contactless fingerprinting was one of the hottest areas of biometric development even before the novel coronavirus was discovered in 2019, along with other contactless biometric technologies in general. The reason is because it allows high scalability leveraging the existing infrastructure such as smartphones with internet connection that we all carry. Adoption and thus further development of such technologies only accelerated during the pandemic. The heat was generated by a series of advances in fingerprint image acquisition using conducive mobile phone camera systems, image processing techniques resulting in high accuracy matching performance with legacy fingerprints and systems, widespread adoption, and recognition from standards bodies.
When we talk about contactless fingerprinting, we mean the acquisition and/or verification of fingerprints without placing the fingers on the type of device we’ve all become used to, for example at border control. Until recently, for a number of reasons, placing your finger(s) on such a reader was the only way of properly acquiring or verifying a fingerprint.
The potential for contactless fingerprint capture technology, particularly through mobile phones, includes a complete reconfiguration of the dominant national ID registration and verification model, Tech5 Co-founder and CTO Rahul Parthe tells Biometric Update in an interview.
This reconfiguration would involve the gradual replacement of a whole generation of dedicated acquisition devices for contact-based fingerprint biometrics. Their replacement would require only software and mobile phones, and the same smartphones will also serve as a face capture devices.
“We all carry these awesome computers in our hands,” Parthe explains. “It’s a perfectly packaged hardware device that is ideal for any capture technology. Smartphones are powerful compute devices on the edge, with a nice integrated camera with auto-focus and flash. And now phones also come with multiple cameras which can help with better focus and depth estimation. This allows the users to take photos of their fingers and the software takes care of the rest. I’d just like to point out here that we’re talking about using the phone’s camera to capture biometrics and using a smartphone to take the place of a dedicated reader. We’re not talking about the in-built fingerprint acquisition we’re all familiar with on many devices which is the means of accessing the device itself.”
The recent state of the art
The technology is coming of age: contactless fingerprinting has reached high accuracy as with contact-based biometric systems when using multiple fingers. A preliminary test conducted by the U.S. National Institute of Standards and Technology (NIST) as of early 2020 proved the same.
Its interoperability with fingerprints captured using contact fingerprint scanners has also been confirmed by NIST, informing guidance released by the agency this past June.
Getting to that point has required innovation in capture technology, Parthe explains. Unlike with a dedicated touch-based device, fingerprint images are captured from varying distances, requiring accurate scale correction to be built into the software.
“There is a lot of machine learning going on with the capture itself,” says Parthe. “To detect the fingerprint, and then once you have detected and segmented, you have to convert it into a format that is acceptable to, let’s say, the national ID solution vendor’s legacy system.”
The images must be adjusted to match reference templates captured with contact-based systems, with 500 dpi, greyscale, and specific contrast characteristics. Only then can the biometric data be processed the same way.
Contact-based fingerprint scanners have advanced dramatically over the past decade, according to Parthe, with smaller and somewhat portable devices coming to market with high accuracy. They still involve a significant investment, however, and require a secondary computing device, like a laptop or tablet, further increasing the overall cost. Also, they limit scalability and require maintenance.
The guidance from NIST is also expected to lead to further documentation, and possibly new standards for contactless fingerprints.
A new generation
Prior to the pandemic, contactless fingerprint solutions were already gaining rapidly in popularity among banks and for digital know your customer (KYC) and remote onboarding applications.
“Major use cases are where people would like to verify themselves against a known reference biometric identity of themselves and do 1:1 verification,” Parthe says.
Contactless fingerprint biometrics adoption is now moving from banks to governments.
Governments are highly concerned with keeping costs down, and reluctant to invest in expensive new scanners, but also have a mandate, one increasingly shared by banks, to reach all citizens. Contact-based biometric scanners typically operate in fixed deployments, making it hard to fulfill that mandate. These systems are also often hindered in bridging the last mile to remote citizens or customers with limitations in internet connectivity and other infrastructure.
The technology best suited to meeting those goals, Parthe says, is provided by the mobile device-based approach. The software now provides similar biometric accuracy to dedicated scanners, he says, with convenience and scalability that opens up new applications.
The software solution is easy to integrate and use, he says, and reduced maintenance compared to hardware-based approaches further adds to the cost benefit, which in turn further improves scalability.
“Especially when you talk about contactless fingerprinting, the fundamental requirement is that somebody has a reference database,” Parthe explains, “which is why the governments who have already collected fingerprints as part of their national ID databases are the ones who are quickly jumping on the use of contactless technologies.”
Shortly after this conversation, Ethiopia announced a trial of Tech5’s technology for its national ID system.
The pandemic provided contactless fingerprinting technologies with “a booster for adoption” according to Parthe, due to concern around the “super-spreader” potential of surfaces meant to be touched by many different people, like traditional biometric scanners.
The T5-AirSnap Finger solution, launched by Tech5 in September, provides an alternative through an SDK that can be easily integrated in applications.
Using software and mobile phones removes the barriers that have prevented biometric technology from being deployed in a range of situations, Parthe points out.
“It basically caters to all the needed and good-to-have features that you want in an acquisition technology.”
Parthe would not be surprised to see governments actually using the technology for enrolling citizens in national ID databases in the near future, he says.
As that happens, he predicts contactless technology will soon start to replace contact scanners. There is also a significant development in the field of fingerprint matching using neural network (NN) based algorithms which have higher tolerance for variation in scale and quality. Combination of the two will significantly increase the effectiveness of these large programs making contactless capture technology an easy choice.
Tech5’s contactless fingerprint offering is called T5-AirSnap Finger. Another available contactless capture technology of Tech5 is called T5-AirSnap Face, and the company’s technology roadmap also includes the development of T5-AirSnap Iris – again for use on a standard mobile phone. Also, with voice biometrics capture the most widely available capture modality for mobile phones, they can effectively provide a complete multi-modal biometrics capture solution, Parthe argues.
For national programs that have traditionally involved SLR cameras, iris scanners, fingerprint scanners, laptops and more, in some cases rolled out to chains of remote islands, mobile phones could provide a powerful alternative.
Parthe notes that NIST’s active involvement in defining the potential certification standards shows they believe in the longer-term potential of contactless fingerprint biometrics.
In the meantime, Tech5’s T5-AirSnap Finger is already serving a growing number of customers for a widening range of applications.
biometric enrollment | biometrics | biometrics at the edge | contactless biometrics | fingerprint readers | machine learning | research and development | SDK | smartphones | T5-AirSnap Finger | TECH5