Deep learning based Detection of One Way Traffic Rule Violation of Three Wheeler Vehicles

Abstract

Nowadays the number of vehicles on the road is increasing incrementally. As a result the number of road accidents is also increasing. The main causes of these accidents are due to violations of traffic rules, high speed, driving in wrong direction etc. Currently monitoring of such cases is done manually. Hence the traffic management and law enforcement has become difficult since it requires tracking of each and every vehicle on the road. In this scenario automated video surveillance system is essential where it provides the capability of automatically detecting security incidents or the abnormal events occurring within the view of camera. This paper proposes a system which automatically detects the one way traffic rule violation without the help of human being. We have considered three wheeler vehicles as they have larger tendency to violate one way traffic rule. Deep learning approach can be used to detect three wheeler vehicles and to identify the one way traffic rule violation. The trajectory points of the vehicle are used to determine direction to which the vehicle is moving. If the direction is opposite to the one way traffic direction, then it can be considered as rule violation.