AI Enabled Home Security System Using Object Detection and Face Recognition with Haar Cascade Algorithm

In this paper smart lock and alert system is implemented using Haar Cascade algorithm. Both face detection and recognition are done by using Haar Cascade algorithm. To share the information with the owner of the house or concerned people, internet of thing (IoT) is incorporated. For all these things to implement effectively, the operating system (OS) based system on chip(SoC) that is Raspberry Pi 4 running Haar Cascade algorithm for face detection and recognition is utilized to constitute edge computing. In this system when the known face is detected, the algorithm gets activated in such a way that door will get unlocked. Additionally, when the unknown face is detected, the algorithm activates, and mail will be sent with an unknown photo as attachment to the authorized person. This enables the owner to take proper action as soon as possible before any danger happens. The algorithm is implementation using python script on Raspberry Pi 4 connected with camera to capture the face at the door area. The key findings from the implementation are accuracy, response time and reliability of model. The algorithm along with a well-maintained database accurately identifies individuals and distinguishes between known and unknown faces. The efficient processing capabilities of Raspberry Pi swift response time since its edge computing rather than cloud computing. The system consistently performs well under different lighting and environmental conditions.