Handbook of Fingerprint Recognition

This fully updated third edition provides in-depth coverage of the state-of-the-art in fingerprint recognition readers, feature extraction, and matching algorithms and applications. Deep learning (resurgence beginning around 2012) has been a game changer for artificial intelligence and, in particular, computer vision and biometrics. Performance improvements (both recognition accuracy and speed) for most biometric modalities can be attributed to the use of deep neural networks along with availability of large training sets and powerful hardware. Fingerprint recognition has also been approached by deep learning, resulting in effective and efficient methods for automated recognition and for learning robust fixed-length representations. However, the tiny ridge details in fingerprints known as minutiae are still competitive with the powerful representations learned by huge neural networks trained on big data. is full professor in the Department of Computer Science (DISI) at the University of Bologna, where he also co-directs the Biometrics Systems Laboratory (BioLab). is full professor in the DISI and a co-director of the BioLab. is university distinguished professor in the Department of Computer Science and Engineering at Michigan State University. is associate professor in the Department of Automation at Tsinghua University.