Suprema’s BioStation 3 gets biometric PAD level 1 conformance stamp from iBeta

Suprema’s facial authentication access control product, BioStation 3, has achieved conformance with level 1 of the ISO/IEC 30107-3 Biometric Presentation Attack Detection standard, following testing by NIST/NVLAP-accredited lab iBeta Quality Assurance [NVLAP Testing Lab Code: 200962] .

A letter of confirmation from iBeta says “testing was conducted with Suprema’s BioStation 3 device, running Firmware v1.3.1 and supported by algorithm FAS_IR_SP v1.0.0. iBeta conducted passive liveness detection testing on the biometric solution from 15 July to 23 July 2024.”

Per ISO/IEC 30107-3 PAD Level 1 compliance guidelines, “testing was conducted in accordance with the contract for a level of spoofing technique that only utilized simple, readily available methods to create artefacts of the genuine biometric for use in the presentation attack.” For level 1 biometric PAD detection standards, iBeta uses only “cooperative subjects” – volunteers who willingly pose for photos, record videos or sit for a live facial cast to have their biometrics collected – and test time for each PAD test per common species of attack (PAI), such as spoofs or masks, is limited to eight hours.

“The test method involved enrolling subjects and having them authenticate five times successfully,” reads the letter. “Six species of presentation attacks (PAs) were then attempted ten times each per subject. A total of 360 presentation attacks were attempted.” At the end of the process, subjects again authenticated five times, to make sure the system still successfully recognized them.

Ultimately, iBeta confirms that “iBeta was not able to gain unauthorized access with the PAs, yielding an overall Presentation Attack (PA) success rate of 0 percent, which then equates to the overall combined Imposter Attack Presentation Accept Rate (IAPAR) of 0 percent on the BioStation 3 device.”

A blog from Suprema announcing compliance says the biometric access control company’s presence in more than 140 countries worldwide has enabled it to “accumulate experience with various racial and ethnic profiles, as well as different facial changes, which is essential for the development of AI algorithms.” It notes the 2022 integration of an advanced neural processing unit into BioStation 3, enabling faster facial authentication “for faces with masks, glasses, hats, beards, niqabs and different hairstyles.”

An extensive explainer on iBeta’s website covering PAD detection and the ISO/IEC 30107-3 standard notes that, while it initially used the term “certification,” an NIST/NVLAP audit in March 2019 corrected the term to “compliance”.

“There is no difference in our test methods, procedures, or processes between the earlier testing and the testing after that date with the exception that the allowed limit for genuine or Bona Fide presentations is now stricter,” it says.

Article Topics

access control  |  biometric liveness detection  |  biometrics  |  face biometrics  |  facial authentication  |  iBeta  |  ISO/IEC 30107-3  |  passive facial liveness  |  presentation attack detection  |  Suprema