Improved features for fingerprint identification

Fingerprint matching is currently the subject of intense research by both private and academic institutions. So fingerprints are emerging as the most common and trusted biometric for personal identification. For the fingerprint matching problem, the input is some fingerprint images and the output is the probability that the fingerprints were captured from the same finger. This paper presents an improved feature for fingerprint identification, it based on the Euclidian distance between the center point and their nearest neighbor bifurcation and ridge ending minutiae´s. These new works overcome the problem of geometric rotation and translation over the acquisition phase of image fingerprints. Fingerprint matching results from the proposed method are validated and the similarity score for the test data available is evaluated. We have tested this technique on more than 1000 images fingerprint taken from “FVC2004”. The experimental results demonstrate the superiority, the effectiveness and the robustness of the proposed method.