Visibility Analysis for Video Surveillance

Now a day's video surveillance is used widely to monitor and record events in shops, galleries, campuses, parking areas, as well as national borders. The objective of this project is to develop a visibility application which determines the number of cameras and their position, in order to cover a campus whose floor plan is available, which leads the reduction in the cost of camera deployment system for video surveillance. While campuses can be of any shape and size, the current work focuses on simple polygonal shapes.

The project is based on Art Gallery Theorem and Triangulation. In the first method, 2-Ear triangulation theorem and 3-color models are used to find the number of cameras and their positions of a polygon having N vertices. The number of positions should be at most. In the second method, to improve the computational speed of triangulation, the polygon is partitioned into monotone pieces and then each monotone is triangulated in linear time. The development of this application is carried out in C/C++ programming language. Software for this project is developed by using Qt.