Secure biometric system for accessing home appliances via Internet

Abstract

Alongside multitude of innovations and consequential changes in lifestyle, the ability to access home appliances or security cameras over the Internet has been in demand for some years now. In this paper, we propose an efficient, low-cost and user-friendly method to access networked home appliances over the Internet, providing strong, well integrated, three levels of security to the whole application and user data. We present a scheme in which, iris image is used for user authentication and after performing its hashing (using BioHashing), it is sent to the server through a secure communication channel utilizing Secure Socket Layer (SSL). The deterministic feature sequence from the iris image is extracted using 1D log-Gabor filters. Mersenne twister random number generator algorithm is used to produce an array of pseudo-random numbers which is later orthonormalised employing GramSchmidt orthonormalization algorithm for performing BioHashing. In addition to this protected interaction mechanism, we use a single Authentication Server in order to enable access to the home appliances of a complete housing society via Internet, which reduces user responsibility and improves accessibility of this endeavor without compromising on system security. We demonstrate the perfect recognition efficiency of this system with equal error rate (EER) of 0% on CASIA 1 iris image dataset.