Enhancing the Security of Financial Transactions using Biometric Authentication and Multi-Task Deep Hybrid Networks

Online banking has become more risky due to the alarming increase in mobile fraud in recent years. Many individuals are wary of using mobile banking due to security concerns, despite its convenience. Secure financial transactions require strong authentication. This study introduces a biometric identification system that uses a mix of fingerprint and face recognition. As the system prepares fingerprints, it uses a Minutiae Matcher and an equalization histogram. In order to extract facial features, PCA is used. Integrating MTDHN predictions with future sensor signal estimates improves forecast accuracy. The MTDHN enhances biometric authentication by finding long-term correlations in sensor data. Values of 97.5 percent for the net present value and 95.15% for the present value demonstrate that the suggested model is an appropriate fit. There is a 97.85% total accuracy rate. All things considered, these outcomes prove that the system can safeguard financial dealings and forestall fraud. Mobile banking should see a dramatic increase in usage as a result of the study's emphasis on biometric authentication, which provides a safe answer to consumers' worries.