Face Recognition for Industrial Shop Floor Access Control

Access Control is an essential defensive tool businesses can use to satisfy both the legal requirements for data privacy and modern security concerns. It is intended to stop unauthorized people and systems from accessing protected resources outside their authorization's scope. The current survey seeks to provide an overview of cutting-edge Access Control methods while showcasing current research trends in this field. The proposed approach is to design a system for authorized employee identification via deep learning algorithms. The algorithm detects the person based on different parameters generated while identifying the face. Generating the number of facial features based on the databases provided by concerned organizations. After detecting the authorized person's face, the entry point will give access to that person. If anyone tries to get access above certain limitations the alert will be generated and sent to the concerned authority with the person's face captured with a camera. For the implementation of the proposed system, after working on the various algorithms, we found out that Haar cascade and MTCNN have their drawbacks while we can use DNN for our proposed system. The project's primary goal is to work on a better algorithm using computer vision techniques that will improve the existing efficiency of the system for facial identification.