Fingerprint recognition algorithms from Neurotechnologija achieved top honors for reliability in this year’s Fingerprint Verification Competition (FVC2006). The VeriFinger 5.0-based algorithm outperformed all other algorithms in the Open Category tested at the average zero False Match Rate (FMR), the most realistic benchmark for evaluating the reliability of real-world biometric systems. The algorithm also won four gold, two silver and two bronze medals in the Open Category. In the Light Category, Neurotechnologija’s algorithm took second place for average zero FMR and won an additional gold and four bronze medals.
“We were very pleased to achieve some of the highest results for the fourth time in a row in the FVC competition,”said Algimantas Malickas, CEO of Neurotechnologija. “While there are many ways to measure the quality of an algorithm, we were pleased that our algorithm ranked number one in the scenario that is most adequate for customers who are implementing real biometric systems. This tells us the new version of our product will continue to provide the high levels of speed and reliability our customers require,” Malickas added.
Organized by the Biometric Systems Lab at the University of Bologna, the Pattern Recognition and Image Processing Laboratory at Michigan State University and the Biometric Test Center at San Jose State University, the FVC competition is held every two years to evaluate fingerprint recognition algorithms from around the world. Neurotechnologija’s algorithms have consistently shown some of the best reliability results in the past four competitions (FVC2000, FVC2002, FVC2004 and FVC2006) and have won a total of 10 gold medals, 12 silver medals and 9 bronze medals.