作者: Tsai-Yang Jea , Viraj S. Chavan , Venu Govindaraju , John K. Schneider
DOI: 10.1117/12.542338
关键词: Data mining 、 Fingerprint recognition 、 Perspective (graphical) 、 Vulnerability (computing) 、 Minutiae 、 Brute-force attack 、 Matching (statistics) 、 Word error rate 、 Engineering 、 Fingerprint
摘要: Despite advances in fingerprint identification techniques, matching incomplete or partial fingerprints still poses a difficult challenge. While the introduction of compact silicon chip-based sensors that capture only part area have made this problem important from commercial perspective, there is also considerable interest on topic for processing and latent obtained at crime scenes. Attempts to match using singular ridge structures-based alignment techniques fail when print does not include such structures (e.g., core delta). We present multi-path approach utilizes localized secondary features derived relative information minutiae. Since minutia-based representation, an ANSI-NIST standard, our has advantage being directly applicable already existing databases. analyze vulnerability systems brute force attacks. The described been tested one FVC2002’s DB1 database11. experimental results show achieves equal error rate 1.25% total 1.8% (with FAR 0.2% FRR 1.6%).