(Received 23 December 2014; accepted 1 December 2015)
Published Online: 27 February 2017
| ||Format||Pages||Price|| |
|PDF (1.9M)||13||$25||  ADD TO CART|
Cite this document
Biometrics deals with authenticating a person's identity based on the physiological or behavioral characteristics. Visual cryptography (VC) is a promising information security technique that allows the secret sharing of images without any cryptographic computations. Various existing schemes were introduced for securing the raw biometric data and template in the database using the VC technique. The complexity of encryption plays a vital role in security improvement. In order to overwhelm the above limitations, a secure authentication management for polar iris templates is presented using VC technique. The collaborative splitting of pixels in all directions presented in this paper was done in order to improve security. At first, the input image was segmented using the Canny edge detection and Hough transform. Subsequently, the normalization module transformed the iris texture from the Cartesian to polar-coordinates. The polar iris image was further separated into two shares, namely, share 1 and share 2, using VC technique. To accomplish more security than the existing methods, both 2-discrete wavelet transform (DWT) and advanced encryption standard (AES) shifting techniques were introduced in VC, termed as transform based AES (TAES). After receiving the encrypted image, the feature extraction is carried out by multi-scale local binary pattern (MLBP). The share 1 images are stored in the user database, whereas the share 2 images are stored in the server database. K-NN classifier is employed to recognize and retrieve the share 2 from the user database on the basis of features. Finally, reconstruction was performed from recognized share 1 and share 2 images by using the inverse process of TAES. The experimental results exhibit better peak signal to noise ratio (PSNR), mean square error (MSE) and normalized correlation (NC), false acceptance rate (FAR), false rejection rate (FRR), and equal error rate (EER) than the other existing methods.
Department of Information Technology, Kings College of Engineering, Gandarvakottai Taluk, Pudukkottai, Tamil Nadu
Arputha Vijaya Selvi, J.
Department of Electronics and Communication Engineering, Kings College of Engineering, Gandarvakottai Taluk, Pudukkottai, Tamil Nadu
Stock #: JTE20140528