Daon and Innovatrics each ace biometric PAD tests by iBeta
Daon and Innovatrics have each successfully completed independent ISO/IEC 30107-3 biometric presentation attack detection (PAD) testing by iBeta Quality Assurance with a 100 percent success rate at identifying spoof attacks, based on a bona fide classification error rate (BPCER) set at 15 percent.
The level 1 test of Daon’s technology shows the compliance of the DaonFace 126.96.36.199 server to the ISO standard.
Innovatrics’ Digital Onboarding Toolkit (DOT), version 4.0 for Android and 2.2 for iOS, was confirmed compliant with level 2 of the PAD standard.
iBeta has been taking steps to enhance the transparency of its testing, and to improve market understanding of what it means.
Daon explains training and detection methods
Daon’s IdentityX platform uses several PAD algorithms to protect against biometric spoofing, according to a company blog post, by Daon Principal Research Scientist Dr. Mike Peirce. The company’s 3D face liveness, its newest PAD solution, applies similar technology to that tested by iBeta, but on a mobile device, with the user moving the device closer to and further away from their face to enable the detection of image distortions that occur naturally during the process.
Peirce explains how Daon built a testing base of 50,000 videos, 40 percent of which were genuine and the other 60 percent biometric spoof attacks. The videos include a wide range of real-world conditions and uncontrolled environments, a range of user devices, and representation from different demographics.
The limits of laboratory testing are discussed in the post, with Peirce urging customers to look for both laboratory confirmation and a track record of customer success from their face biometrics provider.
Innovatrics claims high completion with passive liveness
The passive liveness technology found in Innovatrics’ biometric DOT operates completely on the user’s device.
The company suggests that with lower abandonment rates, passive liveness can deliver onboarding completion rates of 99 percent, compared to 63 percent for active liveness detection.
Innovatrics Head of Global Sales Peter Martis wrote in a recent Biometric Update guest post about the potential of face biometrics for rapid adoption in a wide range of applications.