Abstract: Alteration may be encountered with other biometric modalities which includes face, iris and fingerprints etc. The problem of alteration is especially significant in the case of fingerprints and this is due to the extensive deployment of AFIS in government applications and the simplicity with which fingerprints can easily be complicated. So, here developed an algorithm to automatically detect alteration in fingerprints which is based on the characteristics of the fingerprint features like orientation field and minutiae distribution. The proposed algorithm satisfies the three essential requirements for alteration detection algorithm. They are fast operational time, high true positive rate at low false positive rate, and ease of integration into AFIS (Automated Fingerprint Identification Systems). Fingerprint obfuscation refers to the deliberate alteration of the fingerprint arrangement by a distinct for the purpose of masking his /her identity. Several cases of alterations have been reported. Orientation field describes the ridge movement of fingerprints. Good quality fingerprints offers smooth orientation fields but poses exception near the singular points (core and delta). Gradient-based method is used for computing the orientation field. Mainly it uses the difference between the observed orientation field extracted from the image of fingerprint and the orientation field estimated by the model. The estimated value is represented as a feature vector for classifying the available fingerprint as natural or altered one. In addition, feature extracted from minutiae distribution can also be used for this purpose. Here, minutiae density map is firstly constructed using uniform kernel function and feature vectors from both orientation field discontinuity map and the minutiae density map are shared by concatenating local histograms in each cell which is then fed into a support vector machine (SVM) for further classification.
Keywords: Alteration, Fingerprint obfuscation, NFIQ.
Title: ANALYSIS AND DETECTION OF ALTERED FINGERPRINT MATCHING
Author: Nadhiya Nazeer Khan
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
