Finger print recognition based on biometric cryptosystem

Abstract

In this article, we design a new Biometric Cryptosystem based on finger print recognition. In this Cryptosystem the algorithm is based on Eigen space approach. This approach has been compared with the conventional approaches which are based on exercise object geometry. The execution described uses the arrangement of Eigen space, and by calculating Eigen values and Eigen vectors of the image set. The image set is defined by taking different positions of images of finger prints. For an anonymous input image, the identification algorithm projects this image to the Eigen space and calculates Euclidean distance between the unknown image and the Eigen space. If this distance is minimum then the person is known. Otherwise it is unknown. Experimental results for different number of persons are shown to verify the viability of the designed method.