Mansourehsadat Mousavi - Mousavi
Hosein Jafarkarimi - Department of Computer ، Damavand Branch ،Islamic Azad University، Damavand ، Iran

In this paper, we make an effort to present a method based on Data Mining techniques to distinguish fake and live fingerprints in highly accurate authentication systems. In addition, the types of fraud material that are used in the manufacture of fingerprints was considered so that Based on the type of material, be assessed to discover knowledge about the sensitivity of the sensors based on materials. The material used in this study include gelatin, Eco-Flex, wood glue and latex. The reviewed steps in this research are as followed: First, in the feature extraction stage, we have used methods based on statistical formulae in image histogram. In the next step, the Hybrid algorithm based on support vector machine and decision tree were used. According to the results of this study, fake fingerprints which was made with gelatin, with the lowest errorand latex, with the maximum error was recognized.

Biometric, fingerprint, support vector machine, decision tree