Articles tagged with Biometrics
Researchers at University at Buffalo develop automated system to track individuals' likelihood of committing a terrorist act based on biometrics and behavioral indicators. The system aims to identify perpetrators in real-time, providing a numerical score of malfeasance likelihood.
A new study suggests that diagnostic ultrasound can be used for biometric identification by matching paired images of fingerprints based on internal structures. The researchers found a high accuracy rate of 96% among four readers, with radiologists achieving a 100% success rate.
The USC researchers have created two novel CMOS architectures for 'biometric radar' and automobile accident avoidance systems. One chip can detect living victims in rubble by analyzing minute movements caused by breathing and heartbeat, while the other uses a low-cost CMOS chip to detect ultrawideband signals.
Researchers at NIST found that optimal work surface heights of 26 inches for image quality and 36 inches for speed can improve fingerprint capture. Participants preferred using their right hands, which reduced processing time.
NIST's Multimodal Biometric Application Resource Kit (MBARK) enables flexible and reliable multimodal biometric applications by providing standardized middleware for integrating sensors from different manufacturers. This solution reduces complexity and costs, facilitating the development of secure sites and border access.
The MINEX test demonstrated that minutia templates can be used to match fingerprints accurately, especially when using two index fingers. Systems with images and multiple fingerprint pairs achieved the highest accuracy rates, exceeding 99% in some cases.
Biometrics seeks to overcome its limitations by improving sensor quality, image processing, and combining traits for enhanced accuracy. The technology has shown impressive precision but also faces challenges such as small margins of error, which can have significant consequences.
The National Institute of Standards and Technology (NIST) has released a new software program that judges the quality of scanned fingerprints. The software assigns a quality level ranging from 1 for high-quality prints to 5 for unusable prints, helping ensure accurate matches.