Open CCTV: A video analytic platform for CCTV security | |
| Author | Tissera, Ponsuqe Anjana Chandari |
| Call Number | AIT Thesis no.CS-17-01 |
| Subject(s) | Video compression Electronic data processing--Security measures |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. CS-17-01 |
| Abstract | Surveillance cameras and infrastructure such as Video Management Systems (VMSs) and Digital Video Recorders (DVRs) have become a ubiquitous commodities in the present day world. Video streams obtained from these cameras combined with VMSs/ DVRs can be used to implement very effective and affordable security solutions. Most modern surveillance cameras as well as VMSs/ DVRs have built-in video analytic modules, such as motion detection. But the variety of video analytic modules provided in them and their capabilities are usually very limited. Hence, at present, video feeds in surveillance systems are commonly subjected to analysis of skilled human operators to identify incidents compromising safety and security of a given enterprise. Sometimes this video analysis has to be performed nonstop and in real-time which is tedious to perform manually. With the great progress video analytic software has made and the variety of video analytic types available, it is possible to transfer some of this tedious manual analysis to computers. But very often, advanced analytics require higher memory, processing power, and storage demands that cannot be easily provided in embedded systems of cameras or in VMSs/ DVRs. Therefore, advanced video analytics are usually executed on dedicated servers. If there is a middleware platform to integrate video streams obtained form cameras or VMSs/ DVRs with advanced video analytics, modern advancements in video analysis software can be put into use and the tiresome manual work involved in making existing surveillance systems can be greatly reduced. Thus, in this project, I developed a middleware platform to facilitate flexible and dynamic integration of video streams obtained from various video sources such as direct cameras and VMSs/ DVRs with advanced video analytic modules executed as processes on dedicated servers, such that a large-scale, comprehensive security solution can be constructed with less manual work than would otherwise be necessary. Further, the middleware platform provides advanced video stream integration patterns; for example many types of analysis can be performed on a single video stream, or many video streams can be routed to a single video analytic process. |
| Year | 2017 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis : no. CS-17-01 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Dailey, Matthew N.; |
| Examination Committee(s) | Mongkol Ekpanyapong;Guha, Sumanta; |
| Scholarship Donor(s) | Asian Institute of Technology Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2017 |