The success of Automated Fingerprint Identification Systems (AFIS) has lead to an increased number of incidents where individuals alter their fingerprints in order to evade identification. This is especially seen at border crossings where fingerprints are subject to comparison against a watch list. This chapter discusses methods for automatically detecting altered fingerprints. The methods are based on analyses of two different local characteristics of a fingerprint image. The first analysis identifies irregularities in the pixel-wise orientations which share similar characteristics to singular point. The second analysis compares minutia orientations covering a local, but larger area than the first analysis. A global density map is created in each of the analysis in order to identify the distribution of the analyzed discrepancies. Experimental results suggest that the method yields performance fully comparable to the current state-of-the-art method. Further improvements can be achieved by combining the most efficient analysis of the two methods. The promising results achieved in this study are attractive for further investigations. Especially, studies into the possibility of introducing alteration detection into standard quality measures of fingerprints which would improve AFIS and contribute to the fight against fraud.
CITATION STYLE
Ellingsgaard, J., & Busch, C. (2017). Altered fingerprint detection. In Advances in Computer Vision and Pattern Recognition (pp. 85–123). Springer London. https://doi.org/10.1007/978-3-319-50673-9_5
Mendeley helps you to discover research relevant for your work.