Scopus Indexed Publications

Paper Details


Title
A Proposed Approach to Detect Traffic Signs by using Convolutional Neural Network
Author
Md. Mazbaur Rashid, Mr. Narayan Ranjan Chakraborty, Mr. Shah Md. Tanvir Siddiquee,
Email
Abstract

It is very challenging to predict traffic signs by machine without human intervention. Accidents are becoming more frequent today as a result of improper direction. Automated assistance can reduce these mishaps. For the construction of our object (traffic signs) detection model, A CNN (Convolutional Neural Network) model is utilized. With CNN's assistance, machines can quickly recognize objects and forecast the traffic sign automatically. The classification accuracy of this model for 43 different traffic sign types was 96.02% on average. The major goal of this research is to demonstrate to drivers how to drive safely by anticipating all traffic laws and directions through the machine's instructions. In the future, video processing may add. Moreover, it’s crucial to have a large class size, the ability to predict signs in the dark accurately, and proper direction. In the end, it can be anticipated that this study will bring prosperity to the nation.


Keywords
"Image processing , Traffic Sign , Computer Vision , Convolutional Neural Network , TensorFlow , Image Classification"
Journal or Conference Name
Proceedings of the 8th International Conference on Communication and Electronics Systems, ICCES 2023
Publication Year
2023
Indexing
scopus