BANK SAFE SECURITY SYSTEM USING MACHINE LEARNING WITH FACE AND LIVE DETECTION

Ensuring the security of transactions is currently one of the biggest challenges facing banking systems. The use of biometric authentication of users attracts huge sums money from banks around the world thanks to their convenience and acceptance. Especially in an offline environment where face images from ID documents are compared to digital ones selfie. The selfie vs. id comparison has actually been used in some broader areas as well programs today, such as automatic immigration screening. The great difficulty of such the process is to limit the differences between the comparison face images given their different origins. We propose a new architecture for the cross-domain matching problem based on deep features extracted by two well-referenced convolutional neurons Network (CNN). The results obtained from the collected data, called Face Bank, with more accuracy of more than 93%, indicate the power of the proposed head-to-head comparison problem and its integration into real banking security systems.