Determination of Traffic Flow Distribution at Intersections Bases on Computer Vision & Deep Learning Technology
Every year, the growth in the number of vehicles in Indonesia has always experienced a rapid increase. As a result of the rapid growth of these vehicles, of course, it can also cause problems such as density and congestion in traffic. This research attempts to develop a system to read and analyze vehicle traffic and generate traffic data at the intersection. The research was conducted by looking for several video samples at crossroads and then analyzed using Computer Vision and Deep Learning. After the program can accurately detect the flow of vehicles at the intersection, the program will produce more detailed data. Data details will be processed to produce traffic variables at the intersection. In this study, another way to obtain traffic variables at the intersection is by using IFN calibration. In this case, the platforms used are Python 3.8, OpenCV, and Yolo V3, and there is an IFN program whose primary programming language also comes from Python. From this research, it can be seen whether the program or system created can produce traffic variables. If possible, then this program or system can be used in the future to analyze and overcome various kinds of problems and obstacles that exist in traffic, especially at crossroads.