Enriching Biometric ATM operations through Deep learning
International Research Journal of Modernization in Engineering Technology and Science · 2024
Biometric authentication techniques are becoming more and more prevalent in various implementations, relying on fingerprints and facial features of individuals.Although there are numerous facial recognition systems available.Further research is needed to uncover factors that can enhance efficiency and accuracy.Facial and fingerprint identification are crucial in the identification process due to their ability to operate independently without human intervention, unlike certain other biometric methods.This not only demonstrates the immense potential for enhancing security in Virtual ATM transactions, but also sheds light on the reasons behind the significant interest in biometric identification systems.Thus, a proposed framework has been developed to enhance biometric authentication on Virtual ATMs by utilizing features like Facial and Fingerprint recognition.The framework incorporates Live Streaming and Region of Interest, along with Channel boosted Convolutional Neural Networks.Additionally, OTP authentication has been implemented.The framework has been extensively tested through thorough experimentation to yield very promising results.