Flow Analysis of Vehicles on a Lane using Deep Learning Techniques

Abstract

Increased volume of vehicles and non-availability of complete traffic control system are the motivation to develop smart traffic monitoring system. Traffic state estimation is an important component in smart traffic management. The current traffic density estimators from image sources provide estimation in terms volume of vehicles or percentage of area occupied in the lane. But for a more intelligent traffic management this is not the efficient way hence, a fine grained traffic density distribution for different category of vehicles is necessary. Over the entire coverage area of lane with vehicles flow estimator with respect vehicle categorization, lane density identification, traffic flow analysis are referred in this work as dynamic traffic state map. This work proposes a deep learning solution to constructs dynamic traffic state map from traffic feeds.