Real-Time Authentication with A.I

Abstract

Wrongdoing conception in surveillance videos is more challenging than detection. This work introduces the design of a accurate, lightweight and fast variance understanding neural network. AI capabilities can improve the way current security solutions are working today. It is projected that there will be need for security in area of AI to overcome issues like securing AI training data, training pipelines and ML models, also how AI modelled applications can be operated through negotiating with them. How can AI be leveraged to enhance security of critical infrastructure in the landscape of critical information which may contain large amount of critical business data. Now a day’s state sponsored attacks are growing on an alarming rate, due to which government and private organization are more vigilant in protecting critical infrastructure. This can be achieved by creating business application which will solve the headache of securing critical information for NCIIPC. (National Critical Information Infrastructure protection centre) This can be simulating in high end laptop with Intel Core i5 – 9th generation processor, 6 GB NVIDIA GPU GTX 1660 Ti, which has high FPS which we think will work more smoothly with MTCNN model to detect and recognize the face. Also, at various stages authentication function and alert function will work on top of recognition function. This application will prevent the user to enter into critical infrastructure and prevent wrong doing post validating itself through fake ID Card.

Keywords: AI, ML, NCIIPC, MTCNN, Security, IoT, ID Card, Data-Theft, PII, SPDI