Litcius/Paper detail

Intelligent Traffic Management for Emergency Vehicles using Convolutional Neural Network

S. Deepajothi, D. Palanivel Rajan, P. Karthikeyan, S. Velliangiri

202124 citationsDOI

Abstract

The structure of the road network and the rapid growth of urbanization are becoming increasingly complex. The intersection delay is the main factor that affects the productivity of urban road traffic as the bottleneck of traffic growth. A fair signal control system could help relieve congestion on the highways. Suppose the privilege of preference for the emergency is not assured. In that case, the delay in traffic at the collision may increase, which could hardly indicate the reliable, safe and rapid output as a priority of public transport or any emergency vehicle. We have proposed the Convolutional neural network (CNN) based traffic management for emergency vehicles. CNN model is deployed in the Raspberry-Pi. CNN model will accept the video from the traffic road and take quick decision to allow the emergency vehicles. The proposed method improves accuracy over traditional image processing algorithm and reduces cost.

Topics & Concepts

Convolutional neural networkComputer scienceBottleneckIntersection (aeronautics)Traffic congestionEmergency vehicleReal-time computingComputer networkTransport engineeringArtificial intelligenceEngineeringEmbedded systemFire Detection and Safety SystemsVideo Surveillance and Tracking MethodsSmart Systems and Machine Learning