A new approach to fake finger detection based on skin distortion

This work introduces a new approach for discriminating real fingers from fakes, based on the analysis of human skin elasticity. The user is required to move the finger once it touches the scanner surface, thus deliberately producing skin distortion. A multi-stage feature- extraction technique captures and processes the significant information from a sequence of frames acquired during the finger movement; this information is encoded as a sequence of DistortionCodes and further analyzed to determine the nature of the finger. The experimentation carried out on a database of real and fake fingers shows that the performance of the new approach is very promising.