Object Detection Based Security System Using Machine learning algorthim and Raspberry Pi

Abstract

Conventional security systems that use surveillance cameras to monitor the property lacks the ability to notify the security administrator in the event of trespassing. A security camera when used along with a digital video recorder (DVR) is only effective as a source to gather evidence unless the video feed is constantly being monitored by a dedicated personnel. This paper discusses the implementation of a cost effective, intelligent security system that overcomes drawbacks of conventional security cameras by utilizing a machine learning and Viola-Jones algorithm under image processing literature to identify trespassers and multiple object detection in real time. The paper presents the design and implementation details of the intelligent object detection based security system in two different computing environment, MATLAB and Python respectively using Raspberry Pi 3 B single board computer. The security system is capable of alerting the security administrator through email via internet while activating an alarm locally.