Automatic Fingerprint Matching Using Extended Feature Set

Michigan State University will develop a fingerprint system to automatically extract extended features, including pores, dots, ridge contours and minutiae shape and then use them to improve the fingerprint matching performance, both for ten prints and latents. The goal of this biometrics project is to study the statistical properties of extended features, and to empirically demonstrate the performance gain by combining minutiae and extended features using various fusion schemes. This system will lead to an improved tool for use by state and local criminal justice agencies, to combat terrorism and other types of crime, and will be compatible with NIST standards and interoperable with the FBI's IAFIS system. ca/ncf