Neural Network Based Intelligent Traffic System
B. M. Saathvik, Vinayak Gupta, Ayush Kedia, Lov Asawa, Karpagavalli Subramanian · 2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) · 2021
As years pass by, population and vehicular mobility in most of the cities are rapidly growing which leads to a lot of traffic congestion at junctions. This also makes lots of emergency vehicles like Ambulance, Firefighters, etc to stand in traffic snarls which leads to loss of life during the delay of these vehicles. This paper outlines the development of the Intelligent Traffic System (ITS) prototype by providing solutions to these problems which can handle traffic congestion at right time without manual intervention. ITS uses a neural network model and Internet of Things (IoT) unit to calculate the number of vehicles and emergency vehicles in the traffic lanes and then control the traffic signal based on it. A neural model algorithm is designed in such a way that the model is capable of detecting the ambulance from the live video as well as still images. Eventually, the IoT unit will provide the density of the vehicles in each lane and accordingly allocate the timing for signal transition in the following sequence: RED, YELLOW, GREEN, and the cycle repeats.