IncidentNet: Traffic Incident Detection, Localization and Severity Estimation with Sparse Sensing

Prior art in traffic incident detection rely on high sensor coverage and are primarilybased on decision-tree and random forest models that have limited representation ca- pacity, and as a result cannot detect incidents with high accuracy. This paper presents IncidentNet - a novel approach for classifying, localizing, and estimating the severity of traffic incidents using deep learning models trained on data captured from sparsely placed sensors in urban environments. IncidentNet model works on microscopic traf- fic data that can be collected using cameras installed on traffic intersections. Due to the unavailability of datasets that provide microscopic traffic details and traffic inci- dent details at the same time, a methodology is also presented to generate synthetic microscopic traffic dataset that matches given macroscopic traffic data. IncidentNet achieves traffic incident detection rate of 98%, with false alarm rates of less than 7% in 197 seconds on average in urban environments with cameras on less than 20% of the traffic intersections.