Kneron’s access control biometrics pass Fime performance and PAD assessments

Kneron’s has passed assessments for biometric presentation attack detection and performance in a month-long evaluation of its access control technology by Fime.

The Fime EMEA Lab tested Kneron’s Face Recognition Module 1.0 used in its KnDoorLock software against the ISO/IEC 30107 standard for PAD and performance standard ISO/IEC 19795 with images collected using a near-infrared (NIR) camera.

The PAD test was carried out with 6 Alpha level recipes and 6 Beta level recipes for attack instruments, including printed photos, including curved ones, screen replay attacks, paper masks, and also complex multi-layer masks and full 3D-printed head models with hood masks. Kneron’s liveness detection software delivered both an attack presentation classification error rate (APCER) and a bona fide presentation classification error rate (BPCER) of 0.00 percent. The result, Kneron says, is assurance of “ultra-high security.”

The performance testing was carried out with 150 subjects in low light, strong light, and challenging indoor conditions, according to Kneron’s blog post announcement. The company’s biometrics returned a false acceptance rate (FAR) and false rejection rate (FRR) of 0.0 percent, with no capture on enrollment failures.

“Achieving zero-error performance under NVLAP-validated conditions is a major milestone, not only for Kneron but for the future of privacy-preserving AI,” says Kneron Founder and CEO Albert Liu. “At a time when deepfakes and digital identity fraud are rising globally, this certification demonstrates that edge AI can deliver security the public can trust. It gives our partners and customers the confidence to retire outdated, easily spoofed systems and adopt next- generation authentication built for today’s threats.”

The software runs on Kneron’s purpose-built low-power NPU chips.

In addition to NIST’s NVLAP program, Fime is accredited for FIDO Alliance Biometric Component Certification and meets the requirements for biometric security testing for Android devices.

Article Topics

biometric liveness detection  |  biometric testing  |  face biometrics  |  facial recognition  |  Fime  |  ISO/IEC 19795-2  |  ISO/IEC 30107-3  |  Kneron  |  presentation attack detection