Lithuania-based Neurotechnology has announced a new software development kit for low-power fingerprint recognition called FingerCell 3.0.
The solution is designed for small sensors, and can produce a biometric template using under 128KB of memory. Offering the example application of an ARM Cortex-M4 microcontroller, Neurotechnology says it can produce a template in 650 milliseconds based on a 234 x 332 px, 500 dpi image, and that it can match templates in under four milliseconds.
In other words, FingerCell 3.0 can scan and match fingerprints on small devices with small sensors, even if they don’t have much processing power. In a statement announcing the SDK, Neurotechnology listed example applications such as “mobile phones, fingerprint door locks, access control panels, time and attendance systems, handheld payment and point-of-sale terminals”; but it seems fair to say it could generally have Internet of Things applications.
The SDK’s announcement comes quickly on the heels of Neurotechnology’s success in getting its MegaMatcher On Card algorithm past the NIST MINEX III evaluation, with both developments suggesting a rising profile for the company’s biometric algorithm software.
November 29, 2016 – by Alex Perala