Fingerprint Image Encryption By Using Elliptic Curve Cryptography

Fingerprints are meticulous,distinctive,hard to modify and robust,making them as the human identity of the lifetime.In the last decade,usage of automatic fingerprint identification systems is highly increased.Therefore,users need secure method for protecting their own fingerprint data.In recent years cryptographic techniques have been suggested for information security.Integer factorization and discrete logarithm systems are most useful concepts for the existing system.Elliptic curve cryptography suggested by Koblitz as a tool for securing the information.In this,research we propose a new algorithm to encrypt fingerprint image by using Elliptic Curve Cryptography.The propose method composed two parts such as linear and non-linear.Image pre-processing is linear transformation and encryption by elliptic curve goes to non-linear part.Fingerprint image encryption divided into three main sections;· Image pre-processingWe were applied image resizing and Arnold Cat Map as pre-processing techniques.In addition,to increase the confusing effect of fingerprint encryption,we also utilize a chaotic-behaved method called Arnold Cat Map(ACM)for a 2D shuffling of pixel locations in our method.· Dividing blocks,converting to one integer and encodingIf the number of pixels is large then it will take more computational time.On this research,we use block cipher for reducing the computational time and computational space.We combine and convert pixels values in each block into one big integer.Koblitz's encoding method which is used to map each big integer to a point on the elliptic curve.· Encryption and cipher image.We apply formal approach elliptic curve cryptography for general image encryption.The output is the cipher data composed of a bunch of cipher points on for each blocks.Moreover,for the Purpose of illustration,we can generate the cipher image from the cipher data.Furthermore we extended our encryption scheme to faces and iris images.Not only biometric images we proposed our algorithm for transmission and stored of any other important pictures.We demonstrate it by using Lena's image encryption.Experimental results are carried out with detailed analysis to demonstrate that the proposed fingerprint encryption-decryption algorithm is advantageous in several different aspects including efficiency,security and flexibility.Using this algorithm,we achieve a margin of about 0.1%in the test of Number of Pixel Changing Rate(NPCR)values comparing to the-state-of-the-art performances.We conclude that the proposed algorithm has a great performance in both efficiency and security which is highly practical for biometric data storage and transmission through unsecure networks.