A self banking biometric machine with fake detection applied to fingerprint and iris along with GSM technology for OTP
Joyce Soares, A. N. Gaikwad · 2016
The growing direct or spoofing fraudulent attacks of thieves has motivated us to focus our prime concern on the security over money transaction. The accuracy of biometrics in identification is increasing its usage extensively. The method proposed in this paper focuses on how the money transaction in an ATM machine will be secured by providing personal identification by analyzing biometrics like fingerprints and iris patterns which are known for their steadiness and diversity. Use of biometrics provides a paperless banking environment along with the smart ATM access. In this system the samples of the fingerprint and iris along with the registered mobile number of the customer needs to be collected and saved in the database by the banker if the customer is to access the ATM. The actual operation of the system begins when the customer is to access the ATM to make a money transaction. The fingerprint and iris samples will be captured and matched. The system will distinguish between the real legitimate trait and fake self manufactured synthetic or reconstructed samples by comparing it with the samples saved in the database during enrollment. After finding valid samples the system generates a 3 digit code which is received by the customer on his/her registered mobile number. This process is carried out using a GSM modem interfaced with the ARM7. The entered OTP will be checked, after the OTP is found valid the customer is allowed to make further transactions otherwise the account is blocked. The ATM terminal will also be secured from fire and thieves by including a thermistor and a tilt sensor in the system. The experiments were conducted in real time operation by first performing enrollment and then authentication for two individuals.