Face recognition systems have always been the center of attraction in the technology world and there are numerous problems that it can solve. One of which stuck in our mind is the tedious task of bank tellers to constantly type usernames and ask for verification codes and IDs. So to ease their job we came up with an idea which could revolutionize the banking institutions by leveraging the concept of face recognition.
Smart teller is a face recognition system which uses customers facial image to fetch all the required details thereby eradicating the process of manual verification and validation.
We used the deep learning library named dlib to efficiently extract faces from images and developed methods and algorithms to search similar faces in the blockchain. We used blockchain technology to implement decentralized storage architecture which is highly secure because it stores all the information in an encrypted hash.
We ran in a number of problems which were mostly in the domain of blockchain of how to create blocks, implementing the consensus algorithm and proof of work, etc.
We got hands-on experience with deep learning libraries and developed confidence in face recognition domain by implementing a facial recognition architecture. We even developed a simple blockchain which stored confidential information of the bank customers.
We got a good understanding of how blockchain technologies work which can be the backbone for any huge systems. We learned how to work in teams, develop systems which required advanced architectures, etc.
We want to take the application to the next level by incorporating a smart wallet application using blockchain which will explicitly be available to bank customers where they use face recognition as the entry point in the application thereby providing advanced security.
blockchain deep-learning face-recognition python sql