Intelligent Traffic Signal Automation Based on Computer Vision Techniques Using Deep Learning

Traffic congestion in highly populated urban areas is a huge problem these days. A lot of researchers have proposed many systems to monitor traffic flow and handle through different techniques. But the current are not reliable enough perceive signals real-time. Therefore, we aim build system that can efficiently perform real-time environments solve signal automation. Since vehicle detection counting crucial any system, use state-of-the-art deep learning techniques detect count vehicles We then automate timings by comparing on all sides junction. These automated sufficiently reduce improve flow. prepared dataset 4500 images achieved about 91% accuracy training it Faster RCNN.