High-resolution fingerprint recognition: indexing, feature extraction, and comparison

Abstract

Over the last decades, biometric systems have made their way into our everyday lives. They have made an impact in a wide range of applications, from forensics and national identity programs to mobile device security and payment. Among all the biometric traits, fingerprint is perhaps the most widely explored one because of its distinctiveness and permanence. Specifically, level-2 fingerprint features that include fine details of fingerprint ridge endings and ridge bifurcations, collectively called minutiae, have been extensively studied. On the other hand, level-3 fingerprint features, which include very fine details such as sweat pores, incipient ridges, dots, and ridge contours have not received much attention. One of the reasons could be the requirement of fingerprint images of resolution higher than 800 dpi to explore level-3 features. Additionally, higher cost of high-resolution fingerprint sensors may also have played a role in limiting the research on fingerprint recognition primarily to level-1 and level-2 features. However, with advancements in sensor technology, high-resolution fingerprint sensors have be come affordable and are readily available in the market. This has led to a focus shift and several level-3 feature-based fingerprint recognition approaches have recently been developed. In addition to enhancing the recognition performance when com bined with the level-2 features, level-3 features provide a higher level of security as they are difficult to forge. Level-3 features have also been included in the extended feature set for fingerprint recognition. In this context, it is imperative that methods that fully utilize level-3 features are developed for their potential use in the next generation automated fingerprint recognition systems (AFRS).