| Undergraduate Theses | ||
|
||
Abstract: "Security and Surveillance are two vital issues to be addressed in the commercial and defense arenas. Deploying human resources for these tasks is becoming expensive and better alternative strategies are in need. An autonomous video surveillance system would be such an alternative.The aim of this thesis is to develop an autonomous video surveillance system that continuously perceives the action involved in the live video imagery captured by the surveillance camera and alerts the human operator of any suspicious activity, thus minimizing the requirement of human resources. This report presents the basic framework and methodologies adopted by us in the development of the 'VigilCam video surveillance system project'. It deals with the technical approaches that can be adopted for Moving Object Detection, Object Tracking and Activity Perception. A novel method devised for Multiple Object Extraction is also explained. We also show the experimental results of the VigilCam video surveillance system." |
||
|
System Overview : The tremendous technological advancements made in the field of digital video capture and the accompanying reduction in their prices has made it economical to incorporate them in video surveillance systems. These advanced devices coupled with robust software can be effectively used to work as autonomous surveillance systems, thus weeding out the need for expensive human resources. Most of the surveillance cameras currently used in the commercial, law enforcement and military sectors are being used as a "passive" investigation tool. This begets the need for an autonomous proactive video surveillance system. i.e. computer based systems that continuously monitor the live video imagery captured by surveillance cameras for any suspicious activity and alert the human operator accordingly. This report puts forth the framework and methodologies adopted by us in the development of an autonomous video surveillance system called VigilCam. The basic framework or the block diagram of the VigilCam video surveillance system is shown in the Figure below - 
|
||
|
Environment : Microsoft Visual Studio 6.0, MFC, TWAIN, FreeImage, Video for Windows (VFW), Microsoft Windows |
||
|
Project Members : Deepak Roy Chittajallu, Raghava Teja, Sujeet Kuchibhotla, Vijaya Rao Suravajjala |
||
Software tools developed during the course of this work :
|