Fingerprint is one of the first and most popular forms of biometric to be used for personal identification. Matching fingerprint images is an extremely difficult problem, mainly due to the large variability in different impressions of the same finger, also refereed as intra-class variations. The main factors responsible for intra-class variations are displacement, orientation, partial overlap, non-linear distortion, and various noises. At present, fingerprint authentication entails three distinct classes of techniques: minutiae based, correlation based and hybrid based. However, these methods suffer mainly from feature selection and extraction error as a result of inaccurate detection of minutiae due to missed alignment, either by minutiae or reference point. In this book, new techniques are proposed for fingerprint authentication based on statistical analysis of co-occurrence matrices to overcome the drawbacks of previous methods.