Abstract: Over the years, researchers have developed numerous algorithms for pattern recognition, each one being either a novel approach or a significant improvement over the previous work. Also, each new concept for identifying patterns finds uses in many more fields than what it was initially moulded for. Pattern recognition has found indispensable use in human identification based on unique biometric features. A biometric widely used for identification is the fingerprint. In this paper, we propose and explain a new method for minutiae-point-based pattern recognition for fingerprint identification. Comparison between the angles and edge lengths of polygons formed from the minutiae extracted from a fingerprint image is used as the criterion for matching.
Keywords: Point pattern recognition, boundary polygon algorithm, fingerprint recognition, minutiae based matching.