Fake Fingerprint Detection Based on the Pressure and Contour Shape

초록

Fingerprint recognition has become a widely used and highly reliable biometric technology. However, it is easily spoofed by submitting reproduced fake fingerprints made of silicone or gelatin. This paper proposes a novel approach of detecting the fake fingerprints, based on the measurements of finger’s pressure and foreground contour shape. The ridge width of fake fingerprint is generally thicker than that of the corresponding live due to the gap existing between fake finger and mold. This feature which is extracted based on a flexible segmentation algorithm is denoted as a ratio of ridge area and valley area. Moreover, the fake fingerprint images occasionally have irregular foreground contour due to the elasticity of the fake material or the incomplete surface of fake fingers, and this is measured by the Fourier Descriptor. By adopting the LiveDet 2009 DB in this study, the experimental results show that the proposed method can detect the fake fingerprints with distinguishable characteristics such as unusual thick ridges or concave contour shape, combined with the other measurements to improve performance.