Date of Publication :30th November 2023

Abstract: Traffic signs help drivers stay safe while traveling. Each type of sign has a specific function and tells people what to do in specific situations. Many road users don't know what the signs mean which can cause confusion and accidents. This is where pictorial crosswalks come in handy. Traffic sign classification has been identified as one of the most important tasks for automated cars to perform. Key focus in this research is to develop efficient methods for traffic sign classification, which avoids crashes and foster public trust in autonomous vehicles. In this paper, Convolutional Neural Network (CNN) model is used to recognize the 43 different traffic sign recognition benchmark datasets. The proposed CNN model for classification is 16 times smaller than the original VGG-19 Model, yet still maintains 96.62% accuracy on its own test set.

Reference :