Traffic Sign Detection Based on Faster R-CNN in Scene Graph

DOI

10.2991/meees-18.2018.8How to use a DOI?

Keywords

Intelligent transportation, traffic sign detection, convolutional neural network, deep learning, transfer learning

Abstract

The use of intelligent detection and identification software for traffic signs have been an indispensable part of the advancement of transportation systems and networked cars into an intelligent system. As neural networks become more effective in image recognition and classification testing, they are being applied through the use of traffic sign detection and recognition, gradually deepening the quality of research. Founded on the thoughts of deep learning and transfer learning, this paper uses the method of Faster Region-based Convolutional Neural Networks (Faster R-CNN) and the pre-trained neural network model, Alex net, to detect traffic signs in scene graph. Building off the positive reviews of the Alex Net neural network model in image classification recognition, a target recognition neural network model was trained in the framework of Faster R-CNN neural network using scene image data sets. According to the mark on the test data set, the results indicate that using the pre-trained neural network model can quickly build a traffic sign detection model

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Wei Zhao
AU  - Zhiqiang Wang
AU  - Hongda Yang
PY  - 2018/05
DA  - 2018/05
TI  - Traffic Sign Detection Based on Faster R-CNN in Scene Graph
BT  - Proceedings of the 2018 International Conference on Mechanical, Electrical, Electronic Engineering & Science (MEEES 2018)
PB  - Atlantis Press
SP  - 35
EP  - 42
SN  - 2352-5401
UR  - https://doi.org/10.2991/meees-18.2018.8
DO  - 10.2991/meees-18.2018.8
ID  - Zhao2018/05
ER  -