Optimisation of Energy Consumption in Traffic Video Monitoring Systems Using a Learning-Based Path Prediction Algorithm

The number of CCTV video surveillance systems has grown rapidly over the past decade. As CCTV systems are large energy consumers, the problem of optimising the energy consumption of CCTV systems is urgently needed. In this study, we analyse with mathematical models the energy balance consumption for an architecture that implements a path-by-learning prediction algorithm that predicts the path and destination of a mobile in a CCTV network in order to reduce energy consumption. This method significantly reduces the energy consumption of the CCTV system in real time. An experimental system is designed to evaluate the method and experiments are carried out to demonstrate the validity of the method. The experimental results show that the method has not only significantly improved resource use and reduced energy consumption.