Representation, recovery and matching of 3D minutiae template

Contactless 3D fingerprint images can reveal a variety of depth or shape-related features and some of these have been investigated in the literature. The features extracted from the 3D fingerprint images can also be limited by the nature of sensing technique and the image resolution. The acquisition of 3D fingerprint image in Parziale et al (The surround imager: a multi-camera touchless device to acquire 3D rolled equivalent fingerprints. In: Proceedings of ICB 2006, LNCS, vol 3832, 2006, [1]) uses multiple views of fingerprint from different viewpoints and under different illuminations. Such shape from silhouette method to reconstruct contactless 3D fingerprint has not been successful in precisely recovering the 3D fingerprint ridge details and therefore not investigated for 3D fingerprint matching. The 3D fingerprint images acquired from a range of sensing techniques, e.g. structured lighting or photometric stereo, can provide 3D fingerprint ridge–valley depth details. Recovery and matching of only fingerprint depth or height, as attempted in Wang et al. (IEEE Trans Info Forensics Secur 750–760, 2010 [2]) on database of 11 subjects, is not expected to offer accurate results or recover most discriminant information, especially while matching 3D fingerprints from large lumber of subjects.