A Design Methodology of an Embedded Motion-Detecting Video Surveillance System

Abstract

Malaysia urbanization rate has been growing fast due to rapid development and modernization (Economy Planning Unit, 2015) (World Bank, 2015). With the fast growing urban population, the property crimes associated also rises at an alarming rate (United Nations Human Settlements Program, 2007) ( Mohit Mohammad Abdul, Elsawahli H. Mohamed Hassan, 2015) (Ministry of Home Affairs, 2015). Video surveillance system is one of the trusted and efficient security systems to protect people and property against criminal (Varij Ken, 2015). Three problems and issues regarding current video surveillance system are functionality, flexibility and efficiency. The aim of this project is to design and develop an embedded motion detecting video surveillance system and to produce a functioning prototype as the final result to solve these problems. In order to achieve the aim of this project, three objectives are set to be attained. These objectives are to design and develop a motion detection algorithm based on OpenCV library functions using camera as sensor, to process and execute the algorithm using an embedded micro-computer and to compare its processing performance with a PC and to enable wireless user alert using LAN and internet connection. The first objective is achieved as the motion detection algorithm designed and developed based on background subtraction method using camera sensor and openCV library function is proven to functioning well in detecting for motion changes. The algorithm performs successfully both in BeagleBone Black module and PC and is able to deliver outputs required for embedded motion detection video surveillance system. The second objective is achieved as BeagleBone Black (BBB) module, a micro-computer embedded system is used as main processor for the embedded motion-detecting video surveillance system. The BBB module is able to process and execute the motion detection algorithm designed with image processing functions from OpenCV library. The third objective is achieved as embedded motion-detecting video surveillance system is equipped with wireless user alert function using local area network and internet connection. The system is able to send real-time alert to the user via email attached with the image captured and detected as a threat by the system