Automatic detection of Traffic Light in Smart Transportation systems using Machine Learning

Abstract

The goal of this study would be to develop a platform for anticipating complete and rapid traffic flow data. Everything that could also influence the process of traffic on the road was considered part of the traffic situation, including traffic signals, incidents, demonstrations, and sometimes even road repairs that could generate a traffic bottleneck. Drivers or passengers could make an educated choice if we have previous information that would be very close to approximation about all of the above and many other everyday life circumstances that could impact traffic. It also aids the development of driverless vehicles shortly. Traffic data have been growing tremendously in recent decades, and so we have progressed towards big data ideas for transport. Current road traffic forecasting approaches have used some congestion predictive models and seem to be unsuitable for real-world applications. This aspect prompted us to pursue a solution to traffic flow forecasting based on traffic analysis techniques. Since the amount of information generated for the transport network was enormous, effectively forecasting traffic flow was difficult. Humans intended to employ deep learning, genetics, computational intelligence, & learning-based techniques to evaluate massive data for the transport network with a lot fewer complexities in this project. Image Processing techniques can also be used in the identification of road signs, which aids in the appropriate training of automated driving.