Big gain for Neurotechnology in latent fingerprint biometric matching
Neurotechnology has made advancements in its latest submission to the NIST for the Evaluation of Latent Friction Ridge Technology (ELFT). The evaluation demonstrates a three-fold increase in biometric accuracy compared to its previous submissions, positioning the company among the top five providers of latent fingerprint recognition technology.
The company scored a rank-one hit rate of 92.3 percent and showed relatively fast feature extraction in the evaluation.
The evaluation database comprises 1.6 million identity records, including real case samples contributed by various law enforcement agencies. Key metrics used in the evaluation are the false positive identification rate at a given false negative identification rate (FPIR at FNIR) and the false negative identification rate at a given rank (FNIR at Rank).
Of particular note is Neurotechnology’s algorithm’s performance with the Michigan State Police — Distal dataset, achieving an FNIR at FPIR of 0.01. The company claims this achievement ranks the algorithm as the second most accurate submission in its category.
“We are pleased with the advancements in our latent fingerprint technology, which has achieved up to a threefold improvement in the NIST ELFT evaluation,” says Evaldas Borcovas, biometrics research team lead for Neurotechnology.
In the identical evaluation tests using the FBI’s solved dataset #1, the algorithm successfully matched prints in 92.05 percent of searches. Moreover, when set to tolerate a false positive rate of 1 in 100 searches, the system could identify latent prints in about 84.3 percent of cases without manual intervention.
The enhancements in accuracy lead to increased reliability for the algorithm in real-world scenarios, particularly as these algorithms are utilized by law enforcement agencies across different states. The rank in the biometric recognition evaluation gives companies a competitive advantage over others.
Earlier this year, Neurotechnology’s fingerprint verification algorithm achieved top performance in the NIST Proprietary Fingerprint Template III (PFT III) evaluation.
Innovatrics celebrates two decades of biometrics service
Established in 2004, Innovatrics specializes in delivering biometric systems for fingerprint, face, and iris recognition. As the company marks its 20th anniversary, it has provided biometric services to over 600 projects in 80 countries. Innovatrics has also earned recognition for its performance, ranking among the top performers for the NIST ELFT benchmark.
“From our inception, we’ve always had one of the world’s best fingerprint matching algorithms and added more along the way. Our mission always was to use biometrics to help people,” says Jan Lunter, founder and chief executive officer of Innovatrics.
Article Topics
accuracy | biometric identification | biometric testing | biometrics | ELFT | fingerprint biometrics | fingerprint recognition | Innovatrics | NIST
link