A NSGA-II based approach for camera placement problem in large scale surveillance application

Camera placement has huge impact on the performance of large scale surveillance. The quality of the coverage and cost of the camera network make the problem multi-objective. The problem becomes even more difficult when realistic conditions like obstacles, preference coverage and resolution are considered. In this work we propose a framework for optimal camera placement, giving simultaneous consideration to different qualitative aspects using multi-objective genetic algorithm (NSGA-II). We improve our camera-coverage model by including the realistic parameters such as frontal coverage and the optimal range of view of a camera. We develop a novel camera coverage calculation algorithm which ensures that the proposed set-up is applicable to surveillance tasks such as face recognition and object tracking. Developed algorithm has been tested on realistic floor plans and optimal lay-outs have been reported.