Fingerprint Matching through Back Propagation Neural Network

Abstract

Objectives: This paper is focused with original features used in fingerprint matching founded on back propagation neural network. For finding the features in the images, it can be improved using filtering techniques as isotropic and anisotropic. Isotropic can protect features on the input images but can barely progress the dominance of the images. But on the contrary, anisotropic filtering can successfully eliminate noise from the image just when a consistent point of reference is provided. Methods and Analysis: The filters commonly used namely median filter, gabor filter along with anisotropic filters are used for filtering the noises under direct gray scale enhancement .Whereas for an input image, the narrow ridge direction is predictable and the region of interest is positioned. The uniqueness of a fingerprint is exclusively determined by the local ridge characteristics and their relationships. The ridges and valleys in a fingerprint alternate, flowing in a local constant direction. The two most prominent local ridge characteristics are: 1) ridge ending and, 2) ridge bifurcation. A ridge ending is defined as the point where a ridge ends rapidly. A ridge bifurcation is defined as the point where a ridge forks or diverges into branch ridges. Collectively, these features are called minutiae. Minutiae points are extracted during the enrollment process and then for each authentication. In a fingerprint, they correspond to either a ridge ending or a bifurcation. Minutiae are major features of a fingerprint using which comparisons of one print with another can be made. After finding the minutiae points the filters reducing image noises, smoothing, removing some forms of misfocus and motion blur, is in the front step of image processing. Filtering is also used for preserving the true ridge and valley structures. Findings: The digital results of these features are practically feeded as input of the neural network using median filter, gabor filter, anisotropic filter for training function. For fingerprint identification the confirmation part of the system identifies the fingerprint based training show of the network. Novelty and Improvement: To finish the new outcome reveals that the number of accepted sample rate of the proposed method using the three filters which is far better than the existing fingerprint verification system using artificial neural network