A Framework of Fingerprint Scaling

Abstract

Fingerprint scaling refers to the adjustment of fingerprint images to solve the problem of sensor interoperability. In this paper, we present an innovative framework of fingerprint scaling with minimum modification to existing systems. For the purpose of facilitating system configuration, we have developed a series of scaling methods, including scaling factors, graph- and template-based scaling methods. In graph-based scaling methods, we have explored the application of various technologies in estimation of the average inter-ridge distance. In template-based scaling methods, we have developed an estimation method using Delaunay triangulation algorithm. The experiments show that the performance achieved by using this framework is better than that of original system. With a scaling module, the average EER in our experiments drops from 27.78% to 13.89%.

DOI: http://dx.doi.org/10.11591/telkomnika.v11i3.2304