Secure Fingerprint Biometric Authentication Using Deep Learning & Minutiae Verification
Nowadays, there has been a rise in security concerns regarding fingerprint biometrics. This problem arises due to technological advancements in bypassing and hacking methodologies. This has sparked a need for a more secure platform for identification. In this paper, we will be using a deep Convolutional Neural Network (CNN) as a pre-verification filter to filter out bad or malicious fingerprints. Since deep learning allows the system to get more accurate at detecting and reducing false identification by training itself again and again with test samples, the proposed method improves security and accuracy by multiple folds. We will see the implementation of a secure fingerprint verification platform that takes a fingerprint input as an optical input and is pre-verified using google's pre-trained inception model for deep learning applications and then passed through to a minutia based algorithm for user authentication. The results from our proposed model will be compared with existing platforms and based on that conclusions will be drawn.