Traffic sign detection is one of the most important tasks for autonomous
public transport vehicles. It provides a global view of the traffic signs on the road. In
this chapter, we introduce a traffic sign detection method based on auto-encoders
and Convolutional Neural Networks. For this purpose, we propose an end-to-end
unsupervised/supervised learning method to solve a traffic sign detection task. The
main idea of the proposed approach aims to perform an interconnection between an
auto-encoder and a Convolutional Neural Networks to act as a single network to detect
traffic signs under real-world conditions. The auto-encoder enhances the resolution of
the input images and the convolutional neural network was used to detect and identify
traffic signs. Besides, to build a traffic signs detector with high performance, we
proposed a new traffic sign dataset. It contains more classes than the existing ones,
which contain 10000 images from 73 traffic sign classes captured on the Chinese roads.
The proposed detector proved its efficiency when evaluated on the custom dataset by
achieving a mean average precision of 86.42%.
Keywords: Autonomous public transport vehicles, Deep learning, Smart cities, Traffic sign detection.
