Abstract: Nowadays traffic congestion is a serious issue. Traffic congestion is most predominant in metro cities. There are different causes of traffic congestion such as increasing population, rising incomes leading to more vehicles on the road, insufficient capacity of roads to handle traffic etc. Therefore there is a need for optimizing the traffic management system of the city. This paper introduces a new technique using image processing and big data to build a better traffic management system. The system takes CCTV videos installed at various checkpoints as input and converts the video into image frames. Thereafter, background subtraction is performed on these image frames to obtain only objects of relevance. Haar-based cascade classifier is used to detect vehicles and finally vehicle count is performed using parallel computation power of Hadoop.

Keywords: Intelligent traffic system, Big data, transport system, smart city, reduce traffic congestion, vehicle count, video processing.