Publications

Abstract : this paper discovers the possibility of using visual cryptography for reporting the confidentiality to biometric templates. In addition, the contribution of this paper includes a methodology to preserve the privacy of a face database by decomposing an input private face image into two independent sheet images such that the private face image can be reconstructed only when both sheets are simultaneously available. The XOR operator is used to join the two noisy images and fully recover the original template. The proposed algorithm selects the host images that are most likely to be companionable with the secret image based on geometry and appearance. Then random permutation algorithm (RAP) is used to encrypt the each share with individual host image. In previous research, the reconstructed image quality is pitiable and pixel size of the each share is high which increase the storage requirements of the sheets. To overcome this problem, we have to propose JPEG-LS lossless compression standard refining the image quality as well as diminish the pixel size. A analogous procedure is used to de-identify fingerprint images and iris codes prior to storing them in a central database.