Online Detection of Spoof Fingers for Smartphone-Based Applications

The on-line spoof fingers detection algorithm presented here adopting ridge tracing features as input to Support Vector Machine (SVM) classifier. The wavelet transforms the coefficients of ridge data along ridges into sequences. The derived information relates to geometrical distribution of pores on ridges. As a result, it is more effective to distinguish pore information between spoof and live fingers. The features of pore information on several ridges are inputted to typical SVM system to evaluate the proposed system. The results with benchmark database show the improvement on spoof finger detection.