Facial Recognition and Fingerprint-Based Authentication System for a Smart Home Security Using IoT

Facial recognition-based authentication system is one of the important modules of a smart home or a smart office. It can help us to identify members in real time based on labeled images available in the system, and based on authentication it can allow access to the system. And at the same time, it can flag and maintain records at server for non-members visiting home and restrict non-members from entering the home. To protect the system from being fooled by using photos of members, in this research paper, the ToF camera is used with some other strategies and a unique algorithm to make it robust. Implementation of this system requires Raspberry pi 3 as an edge device with ToF camera, DC 12 V door lock, doorbell, relay for connecting these actuators from Pi, Wi-Fi, power bank, a local server for keep storing the images incoming from the edge device and a cloud account for the data analytics. We are using python code and its libraries for image processing and implementation of all the communications systems. Objective of this paper is to present a robust system on facial recognition and fingerprint-based authentication and its implementations in a logical manner to secure a home from intruders and provide members a better security at the same time information about activities outside the home.