Efficient Minutiae Matching Algorithm for Fingerprint Recognition

Recognition of human fingerprint verifies the match among two fingerprints in an automatic way and it is applied in various fields. The fingerprints are unique and its pattern will remain the same for the lifetime. The minutiae points represent the features of fingerprint that aids in the authentication of fingerprints. The main aim of this paper is to improve a scheme for verification of fingerprint by means of feature extraction and matching techniques. The initial step is preprocessing that involves image enhancement and binarization processes for the poor quality input fingerprint images. The fingerprint verification involves two main steps namely minutiae extraction and minutiae matching. The false minutiae points are to be removed and only efficient minutiae points are to be considered for further process. In this work, two publically available fingerprint datasets are utilized and the accuracy of fingerprint recognition is evaluated using the performance measures namely False Matching Ratio (FMR), False Non Matching Ratio (FNMR) and Threshold. From the results, it is clear that our work provides better results in fingerprint recognition.