Smart Traffic Light Switching and Traffic Density Calculation Model using Computer Vision

Abstract

Different Traffic control systems have played a crucial part in traffic management around the globe, especially in densely populated major cities, but they are still not as efficient as they could be. Perhaps some changes can be made to better deal with the traffic in this ever-changing traffic density environment. Traffic congestion has consistently been a rage issue in numerous urban cities. The traditional way was to give each lane a specific predefined time with the green light and had to stop for the rest of the time. Even the lanes with no traffic got the same amount of time as the lane with huge traffic jams. These were promoting traffic congestion rather than solving the issue. Thus, the need for a better system has emerged for changing the current traffic handling setup to be smarter enough to meet this ever-changing demand. In this paper, the idea of traffic lights controlled by live video feed is explored with an enhanced traffic flow system to optimally benefit from the computer vision technology used.