A new template protection approach for iris recognition

Biometric recognition systems, which rely on physical and behavioral features of the human body to recognize a human-being, are used in various areas that require a high degree of security. Among different biometric recognition methods, iris recognition is measured as being the most trustworthy, distinct, consistent, and stable option. Template security is an important aspect of a biometric system. It necessitates the development of an approach that ensures user security and privacy. In this paper, an approach based on steganography is proposed to protect the iris template. Random number based embedding is used in LSB (Least Significant Bit) steganography to enhance security. To provide more security, bits are embedded into LSB's of blue pixel only. The IrisCode bits are embedded across three least significant bits randomly. The resulting template is more secure as original biometric data is not stored in the database rather it is stored after embedding in cover image. The performance of the proposed approach is evaluated and found to be better in terms of Peak Signal to Noise Ratio (PSNR) value, histogram plot and Receiver Operating Characteristic (ROC) curve plot.