Rank One patents passive biometric facial liveness for standard mobile devices
Rank One has been awarded a patent for a passive biometric liveness detection technology based on micro-texture analysis by the U.S. Patent and Trademark Office (USPTO).
The patent for a “Method and system for implementing image authentication for authenticating persons or items” describes an anti-spoofing system that could extract image regions to analyze the spatial relationship between pixels or groups of pixels, comparing the relationship against “a plurality of spatial relationships” characteristic of certain artefacts. An authenticity value is generated based on this comparison.
The authentication methods covered offer several practical advantages over other methods of biometric facial liveness verification, according to the company announcement. Among them, the technology works on standard RGB cameras implemented in mobile phones and other devices, without any special hardware; the passive method does not require any particular action from the user; and the method works with either a still image or a single video frame, for low processing or bandwidth requirements.
Rank One’s ROC facial recognition SDK library has already included the anti-spoofing biometric liveness algorithm for several years, the company says, giving customers a lightweight and easy-to-integrate capability.
The liveness feature is also already integrated into multiple operational access control products. The company also notes that biometric integrators are increasingly deploying facial recognition for contactless payments, access to online banking services and a range of other applications in which anti-spoofing technology is needed to meet security requirements.
According to the announcement, the patent is Rank One’s first, as most of its intellectual property is protected as trade secrets.
Rank One reached a partnership agreement with Infinity Optics last month to provide its algorithms in Infinity’s QuantumCrypt.