A machine vision based working traffic emission estimation and surveillance schema

This paper suggest a new schema for improving accuracy of estimation of traffic CO₂ emissions. The emission estimation is implemented as a part of novel traffic surveillance system which is movable and uses data fusion of several sensors and databases. The system is able to determine the emissions in real-time based on the traffic flow observed and this provides advantages over the current methods. The emissions are often approximated by using estimates of average traffic flow and emission rates but this produces very unreliable results. Another way is to use gas sensors but they are expensive and provide only point measurement data. In this paper, we show the feasibility of the novel schema.