A Novel Approach for Computing Congestion Degree of Road Traffic using MapReduce Framework
In today’s world due to growing population and migration of humans in the urban area, pressure on cities road and road traffic environment has increased exponentially, which leads to a traffic jam situation, waiting on squares, growth in fuel consumption, and increase in travel time from source to destination respectively. Hence there is a need for an effective traffic management system to address the problem of urban area road traffic. The biggest challenge is the collection of road traffic data from various sources such as sensors and video surveillance cameras and processed it in Hadoop Distributed File System (HDFS). In this paper, we have proposed the novel approach for congestion degree computation using the MapReduce framework in the HDFS. The proposed approach is divided into three parts as 1) An efficient framework for road traffic data acquisition using the video cameras, 2) Extraction of road traffic information from surveillance video and 3) Process the traffic data in the HDFS using the MapReduce framework. First, the road traffic data is processed to identify a number of the vehicle, type of vehicle on the road and the speed of vehicles using the vehicle detail extraction algorithm. Second, the extracted information from the video is stored in HDFS using a two-level MapReduce function. That can be used for counting the number of vehicles and computing the congestion degree for that road. Experimental results show that the proposed method successfully process the road traffic data and compute the congestion degree for efficient traffic management.