Performance Evaluation of Fingerprint Enhancement Algorithms

Enhancement of image quality is a key problem for automated fingerprint recognition systems, especially for minutiae-based analysis. In this paper, five enhancement algorithms, including three popular enhancement algorithms and two combined algorithms, were reviewed. The FVC2000 DB1 B was chosen as the testing database. An algorithm evaluating platform based on minutiae comparison was implemented. The performances of the five enhancement algorithms are evaluated in terms of False Match Rate (FMR) and False Non-Match Rate (FNMR) of the fingerprint recognition system. A guidance for the algorithm selection in fingerprint recognition system is drawn from the results.