DEVELOPMENT OF FINGERPRINT BASED AUTHENTICATION FRAMEWORK FOR AUTOMATED TELLER MACHINE

Abstract

The security of the transactions on the Automated Teller Machine (ATM) has been raising many concerns in some parts of the world. These concerns are due to series of limitations in the existing designs of the various service points. The current use of Personal Identification Number (PIN) for ATM user’s verification and identification has been susceptible to unauthorized access, misplacement, forgetfulness, card swallowing among others; thereby narrowing down the machine’s acceptability and patronage. In this thesis, a framework for fingerprint authenticated ATM application is presented. The framework comprises of modules for fingerprint enrolment, database and verification. The verification module comprises of fingerprint enhancement, feature extraction and matching sub-modules, which all rely on suitable mathematical models to function. There is also platform for financial transaction including withdrawal and balance enquiries. The implementation was carried out using Windows 7 as operational platform while C# and Microsoft SQL server served as the frontend and backend engines respectively. Tests conducted on the evaluation of the application using False Rejection Rate (FRR), False Acceptance Rate (FAR) and Average Matching Time (AMT) show the adequacy and the suitability of the proposed framework for ATM user verification and authentication.