ROC.ai scores high mark with fingerprint segmentation algorithm in NIST evaluation
A tenprint fingerprint segmentation algorithm from Rank One Computing has topped the results of an evaluation by the U.S. National Institute of Standards and Technology for accurate segmentation of biometrics, and placed the company in the top 3 worldwide, according to an announcement.
The Slap Fingerprint Segmentation (SlapSeg) Evaluation performed on ROC.ai’s algorithm on July 5 showed 95.7 percent accuracy for detecting 8 fingers with three-inch identification flats. Seven other vendors have scored accuracy between 95.2 percent and 95.6 percent for the same task.
Fingerprint segmentation is a process in which an image of the friction ridge structure of a subject’s hand is separated into individual images of the tops of each finger. NIST’s evaluation also covers two-inch, five-inch and eight-inch scans
The result demonstrates ROC.ai’s status as a developer of advanced biometric software and leadership in the domestic market, the company says.
“We’re elated with the standout results we’ve achieved with this nascent capability,” says ROC.ai CEO Scott Swann. “The commendations we’ve received both domestically and internationally speak volumes about the precision and dependability of our fingerprint segmentation algorithm. It’s a tribute to our team’s incredible skill, commitment, and trailblazing ethos.”
The algorithm tested by NIST is the one provided in ROC SDK v2.5.0.
ROC.ai says it will enhance the SDK this year, along with its live video biometrics and analytics platform, ROC Watch.
Benji Hutchinson was hired by ROC.ai as its chief revenue officer in July.