An efficient algorithm for detecting traffic congestion and a framework for smart traffic control system

Since the number of vehicles is increasing day by day, traffic jams are becoming a common scenario in large cities like Dhaka. These frequent traffic jams at major junctions kill a lot of man hours. Thus it creates a need for an efficient traffic management system. This paper proposes to implement a smart traffic control system which is based on the measurement of traffic density using real time video processing technique. The video sequences from a camera are analyzed using object detection and counting methods to obtain the most effective way. The computed vehicle density is compared with other parts of the traffic in order to control the traffic signal brilliantly. The system has an advantage of using RFID sensors to ensure law enforcement. Therefore, any car or vehicle which breaks the traffic rules can be easily caught. Through this paper we tried to present a progress in the existing manual traffic control system.

I. Introduction

Traffic light posts are positioned at road intersections and pedestrian crossings. Traffic light posts blink the light Signals after a certain time period which is not a complete systematic system [1] as it cannot solve the traffic problems fully. Thus traffic jams take place. Lack of trained traffic police officers [3] and old manual traffic light control system made this problem worse in many cities like Dhaka, Chittagong etc. Since the traffic vehicle pressure is not same at every road at the same time, Traffic lights should be controlled by an adaptive system which will detect the traffic conditions and use traffic light signals accordingly. Instead of using electronic sensors embedded in the road [2] our system is based on Video processing which is a form of signal processing and for which the input is a video frame; the output of video processing may be either an image or a set of characteristics or parameters related to the image. Our system will detect vehicles through video frames instead of using any other Mechanism. A camera will be installed alongside the traffic light. It will capture video sequences of traffic condition on road. This technique will analyse the videos from cameras and count the number of vehicles for each direction and also import to the controller. Then the controller estimates a period of time needed by each path to open and each traffic light to turn on or off based on the number of vehicles in a fixed sequence. It will also be used to monitor the traffic conditions. It can reduce the traffic congestion and avoid the time being wasted by a green light on an empty road. This system is more reliable to detect vehicle presence because it uses actual traffic condition images. This system is intended for country like Bangladesh where traffic policeman can take clever, critical decisions and handle emergencies but with the help of our automatic system they can use preset signal timings to control traffic at intersections because it provides more accurate information for signal decision making.