Secure authentication using image processing and visual cryptography for banking applications

Core banking is a set of services provided by a group of networked bank branches. Bank customers may access their funds and perform other simple transactions from any of the member branch offices. The major issue in core banking is the authenticity of the customer. Due to unavoidable hacking of the databases on the internet, it is always quite difficult to trust the information on the internet. To solve this problem of authentication, we are proposing an algorithm based on image processing and visual cryptography. This paper proposes a technique of processing the signature of a customer and then dividing it into shares. Total number of shares to be created is depending on the scheme chosen by the bank. When two shares are created, one is stored in the Bank database and the other is kept by the customer. The customer has to present the share during all of his transactions. This share is stacked with the first share to get the original signature. The Correlation method is used to take the decision on acceptance or rejection of the output and authenticate the customer. © 2008 IEEE