Fingerprints are the most widely used parameter for personal identification amongst all biometrics based personal authentication systems. As most Automatic Fingerprint Recognition Systems are based on local ridge features known as minutiae, marking minutiae accurately and rejecting false ones is critically important. In this paper we propose an algorithm for extracting minutiae from a fingerprint image using the binary Hit or Miss transform (HMT) of mathematical morphology. We have developed and tested structuring elements for different types of minutiae present in a fingerprint image to be used by the HMT after preprocessing the image with morphological operators. This results in efficient minutiae detection, thereby saving a lot of effort in the post processing stage. The algorithm is tested on a large number of images. Experimental results depict the effectiveness of the proposed technique.
展开▼