Read more
Biometrics such as fingerprint, face, gait, iris, voice and signature, recognizes one's identity using his/her physiological or behavioral characteristics. Among these biometric signs, fingerprint has been researched the longest period of time, and shows the most promising future in real-world applications. However, because of the complex distortions among the different impressions of the same finger, fingerprint recognition is still a challenging problem.
Computational Algorithms for Fingerprint Recognition presents an entire range of novel computational algorithms for fingerprint recognition. These include feature extraction, indexing, matching, classification, and performance prediction/validation methods, which have been compared with state-of-art algorithms and found to be effective and efficient on real-world data. All the algorithms have been evaluated on NIST-4 database from National Institute of Standards and Technology (NIST). Specific algorithms addressed include:
-Learned template based minutiae extraction algorithm,
-Triplets of minutiae based fingerprint indexing algorithm,
-Genetic algorithm based fingerprint matching algorithm,
-Genetic programming based feature learning algorithm for fingerprint classification,
-Comparison of classification and indexing based approaches for identification,
-Fundamental fingerprint matching performance prediction analysis and its validation.
Computational Algorithms for Fingerprint Recognition is designed for a professional audience composed of researchers and practitioners in industry. This book is also suitable as a secondary text for graduate-level students in computer science and engineering.
Summary
Fingerprints are the most established from of biometrics with the most promising future in real-world applications. However, because of the complex distortions among the different impressions of the same finger, fingerprint recognition is still a challenging problem. This book presents an entire range of novel computational algorithms for fingerprint recognition, all evaluated by the National Institute of Standards and Technology (NIST). These include feature extraction, indexing, matching, classification, and performance prediction/validation methods, which have been compared with state-of-art algorithms and found to be effective and efficient on real-world data.
