Protecting Iris Images through Asymmetric Digital Watermarking

When biometric systems require raw images to be stored in centralized databases, it is imperative that appropriate measures are taken to secure these images. A combination of asymmetric digital watermarking and cryptography can serve as a powerful mechanism for facilitating such security needs. The combination of these techniques enables the system to handle many issues associated with storing and using raw biometric data. In this paper, we propose a framework that encodes voice feature descriptors in raw iris images thereby offering an example of a secure biometric system. The contributions of this work are as follows: application of biometric watermarking to iris images in order to provide an added level of authentication; a mechanism to validate the originating source of iris images; understanding levels in which watermarks can be compromised in a biometric system; and implementation of an asymmetric watermarking framework.