Intelligent Surveillance in Smart City Using 3D Road Monitoring

Abstract

Traffic accidents and lack of road monitoring are never-ending cause of tribulation in metropolitan areas. A solution is required to tackle with these problems. The human eye is not capable of capturing each and everything happening on the roads. To build a secure environment in smart cities an intelligent surveillance system is required to keep an eye on what is happening on roads. The current security measures for surveillance are not enough to capture everything. For instance, single view and fixed CCTV cameras are installed on roads. They are only capable of providing information on a fixed angle. The current solution lacks at recognizing the type, model, and license plate of the vehicle. Therefore, the objective of this paper is to develop a 3D road monitoring model to provide intelligent surveillance in smart cities. The proposed solution is capable of performing efficient vehicle detection along with sensing unusual activities on the road. For this, pre-trained models trained using deep neural networks for vehicle detection, license plate recognition, and unusual activity detection are combined through stacking. It will reduce the man work, time, and complexity of the traffic controls.