Fingerprints are the oldest and most widely used form of biometric identification. It consists of patterns which correspond to the ridges and valleys on the fingerprint. Fingerprint images may be degraded and corrupted due to variations in skin and impression conditions. Singular points (SPs) detection is a crucial phase in fingerprint authentication systems and is used for fingerprint classification, alignment and matching. In this book we propose a fault-tolerant approach for detection of genuine SPs especially in noisy fingerprint images. To improve the accuracy, we present multistage detection and elimination of spurious SPs in degraded fingerprint images using three stages. In first stage two different methods, viz., quadrant change and orientation reliability measure, are independently employed on the same image to generate two sets of candidate SPs. The second stage performs the multiscale analysis on a set of candidate SPs located by reliability method. In the third stage, the spurious SPs are detected and thereby eliminated by taking the intersection of the two sets of SPs.