Abstract: The behavioral characteristics are used to identify the person in biometric recognition systems. Every person has different traits which are used to authenticate their identity. Similarly, In the IRIS recognition system, the human eye is used to identify the identity of the individual. But these systems are suffering from the problems of security which is a big concern in recognition. Several mechanisms have been developed by different researchers in last few years that provide protection to the IRIS template. Likewise, a new method has proposed in this thesis, which is acquired by combining the RLE and RSA algorithm together. The RLE provides compression of data and RSA provides encryption of images for security. The proposed method provides high level of security by compressing the template, encrypting the template with RSA and then acquired template is hiding under another cover image. With the application of the proposed technique, template becomes more protected and secure. In order to conclude the efficiency of the proposed approach, experimental analysis has done in terms of different performance parameters such as SSIM, PSNR, MSE, BER and correlation coefficient. The approach has been compared with the traditional LSB technique over 10 different images. From the results analysis, it has been concluded that the proposed RLE-RSA algorithm outperforms the traditional approach regarding efficiency, proficiency and security.

Keywords: Template Protection, RLE algorithm, RSA algorithm, Correlation Coefficient, PSNR.