Human physiological and behavioral characteristics are considered to be more reliable biometric parameters which are utmost protected and authenticated tool for recognizing human identity. Currently, biometric recognition system uses fingerprints as standard parameter and most extensively it act as backbone for human identification. The proposed identification method practices certain details marked on the fingers called minutiae which are the end and crossing points of the ridges rather than using all selective features available in fingerprints. In this work we tried to explore the methods that improve identification accuracy with minutiae matching algorithm using both its local and global structures. Before being evaluated the fingerprint image has been preprocessed using image enhancement as well as histogram equalization method for improving the image quality. In order to have post processing of fingerprint image we have used Fast Fourier Transformation (FFT), binarization, image segmentation, thinning and removal of false minutiae. For fingerprint verification and identification an extensive model and system has been developed using minutiae extraction and matching techniques. An orientation and alignment based minutiae matching algorithm has been used which is proficient in finding correspondence correlation between the input and stored minutiae templates. The proposed model generates the simple, reliable and unique system to identify the person as it uses morphological operations, triple branch counting, decomposing of branch and X-Y coordinate system in matlab which can easily be used to recognize the fingerprint of a person.

Dactyloscopy, basal cells, Biometric, Fingerprints, Minutiae