Traffic system improvisation has been emerging as a societal concern because of increasing number of deaths due to road accidents. This problem highlights the need to study traffic accidents and propose a solution by analyzing the responsible factors. Development of an automated system for controlling the traffic in comparison of traditional methods is the need of today. This could be achieved by analyzing the density of the traffic at any location especially in the cities where the count of accident hotspot is high. This paper proposed CNN-based model that performed hyperparameter tuning of sequential model to classify road traffic conditions base on traffic conditions images collected from Internet. Model classifies the image content and measures the congestion of traffic. Traffic conditions are classified into two levels as high density and low density with an accuracy of nearly 90%.
Road Traffic Density Classification to Improvise Traffic System Using Convolutional Neural Network (CNN)
Lect. Notes in Networks, Syst.
Proceedings of International Conference on Recent Trends in Computing ; Chapter : 34 ; 397-404
2023-03-21
8 pages
Article/Chapter (Book)
Electronic Resource
English
Road traffic target tracking method based on deep convolutional neural network
European Patent Office | 2022
|Traffic Sign Classification with a Convolutional Network
British Library Online Contents | 2018
|INTEGRATED CONVOLUTIONAL NEURAL NETWORKS APPLICATION FOR ROAD TRAFFIC EFFICIENCY
DOAJ | 2024
|