Automatic Generation of Traffic Signal Based on Traffic Volume

Abstract

Now day's computer vision techniques are used for analysis of traffic surveillance videos which is gaining more importance. This analysis of videos can be useful for public safety and for traffic management. In recent time, there has been an increased scope for analysis of traffic activity automatically. Computer based surveillance algorithms and systems are used to extract information from the videos which is also called as Video analytics. Detection of traffic violations such as illegal turns and identification of pedestrians, vehicles from traffic videos can be done by using computer vision and pattern recognition techniques. Object detection is the process of identifying instances of real world objects which include persons, faces and vehicles in images or videos. Object detection is becoming an increasingly important challenge now days as it has so many applications. Vehicle detection helps in core detection of multiple functions such as Adaptive cruise control, forward collision warning. Automatic Generation of Traffic Signal based on Traffic Volume system can be used for traffic control. Traffic Surveillance videos of vehicles are taken as input from MIT Traffic dataset. These videos are further processed frame by frame where the background subtraction is done with the help of Gaussian Mixture Model (GMM). From the background subtracted result some amount of noise is removed with the help of Morphological opening operation and Blob analysis is done in order to the detect the vehicles. Later the vehicles are counted by incrementing the counter whenever a bounding box is appeared for the detected vehicle. Finally a signal is generated depending on the count in each frame.