Lane Line Detection based on Mask R-CNN

DOI

10.2991/icmeit-19.2019.111How to use a DOI?

Keywords

lane detection; deep learning; the diversity of samples; Mask R-CNN.

Abstract

Lane detection plays an important role in driverless system. However, the complexity of the actual road environment makes lane detection more challenging. In recent years, the rapid development of deep learning has pointed out the direction to solve this problem. Deep learning does not care about the change of environment, but only about the diversity of samples. As long as enough samples are trained, the target can be detected and identified. Based on this, a lane detection algorithm based on Mask R-CNN is proposed, which can not only detect lane quickly, but also reach to a total 97.9% of accuracy on our TSD-Max datasets.

Copyright

© 2019, 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  - Bin Liu
AU  - Hongzhe Liu
AU  - Jiazheng Yuan
PY  - 2019/04
DA  - 2019/04
TI  - Lane Line Detection based on Mask R-CNN
BT  - Proceedings of the 3rd International Conference on Mechatronics Engineering and Information Technology (ICMEIT 2019)
PB  - Atlantis Press
SP  - 696
EP  - 699
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmeit-19.2019.111
DO  - 10.2991/icmeit-19.2019.111
ID  - Liu2019/04
ER  -