In this paper, a deep learning based road traffic signs recognition method is developed which is very promising in the development of Advanced Driver Assistance Systems (ADAS) and autonomous vehicles. The system architecture is designed to extract main features from images of traffic signs to classify them under different categories. The presented method uses a modified LeNet-5 network to extract a deep representation of traffic signs to perform the recognition. It is constituted of a Convolutional Neural Network (CNN) modified by connecting the output of all convolutional layers to the Multilayer Perceptron (MLP). The training is conducted using the German Traffic Sign Dataset and achieves good results on recognizing traffic signs.
Traffic signs recognition with deep learning
01.11.2018
1152591 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Classification and Recognition of Traffic Signs Using Deep Learning
Springer Verlag | 2022
|Wiley | 2022
|DEEP LEARNING BASED TRAFFIC SIGNS BOUNDARY ESTIMATION
British Library Conference Proceedings | 2020
|Deep Learning Based Traffic Signs Boundary Estimation
IEEE | 2020
|Traffic Signs, Visibility and Recognition
British Library Conference Proceedings | 1999
|