IEICE TRANSACTIONS on Fundamentals
Fast Lane Detection Based on Deep Convolutional Neural Network and Automatic Training Data Labeling
Summary :
Lane detection or road detection is one of the key features of autonomous driving. In computer vision area, it is still a very challenging target since there are various types of road scenarios which require a very high robustness of the algorithm. And considering the rather high speed of the vehicles, high efficiency is also a very important requirement for practicable application of autonomous driving. In this paper, we propose a deep convolution neural network based lane detection method, which consider the lane detection task as a pixel level segmentation of the lane markings. We also propose an automatic training data generating method, which can significantly reduce the effort of the training phase. Experiment proves that our method can achieve high accuracy for various road scenes in real-time.
- Publication
- IEICE TRANSACTIONS on Fundamentals Vol.E102-A No.3 pp.566-575
- Publication Date
- 2019/03/01
- Publicized
- Online ISSN
- 1745-1337
- DOI
- 10.1587/transfun.E102.A.566
- Type of Manuscript
- PAPER
- Category
- Image
Authors
Xun PAN
Waseda University
Harutoshi OGAI
Waseda University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Xun PAN, Harutoshi OGAI, "Fast Lane Detection Based on Deep Convolutional Neural Network and Automatic Training Data Labeling" in IEICE TRANSACTIONS on Fundamentals,
vol. E102-A, no. 3, pp. 566-575, March 2019, doi: 10.1587/transfun.E102.A.566.
Abstract: Lane detection or road detection is one of the key features of autonomous driving. In computer vision area, it is still a very challenging target since there are various types of road scenarios which require a very high robustness of the algorithm. And considering the rather high speed of the vehicles, high efficiency is also a very important requirement for practicable application of autonomous driving. In this paper, we propose a deep convolution neural network based lane detection method, which consider the lane detection task as a pixel level segmentation of the lane markings. We also propose an automatic training data generating method, which can significantly reduce the effort of the training phase. Experiment proves that our method can achieve high accuracy for various road scenes in real-time.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E102.A.566/_p
Copy
@ARTICLE{e102-a_3_566,
author={Xun PAN, Harutoshi OGAI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Fast Lane Detection Based on Deep Convolutional Neural Network and Automatic Training Data Labeling},
year={2019},
volume={E102-A},
number={3},
pages={566-575},
abstract={Lane detection or road detection is one of the key features of autonomous driving. In computer vision area, it is still a very challenging target since there are various types of road scenarios which require a very high robustness of the algorithm. And considering the rather high speed of the vehicles, high efficiency is also a very important requirement for practicable application of autonomous driving. In this paper, we propose a deep convolution neural network based lane detection method, which consider the lane detection task as a pixel level segmentation of the lane markings. We also propose an automatic training data generating method, which can significantly reduce the effort of the training phase. Experiment proves that our method can achieve high accuracy for various road scenes in real-time.},
keywords={},
doi={10.1587/transfun.E102.A.566},
ISSN={1745-1337},
month={March},}
Copy
TY - JOUR
TI - Fast Lane Detection Based on Deep Convolutional Neural Network and Automatic Training Data Labeling
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 566
EP - 575
AU - Xun PAN
AU - Harutoshi OGAI
PY - 2019
DO - 10.1587/transfun.E102.A.566
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E102-A
IS - 3
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - March 2019
AB - Lane detection or road detection is one of the key features of autonomous driving. In computer vision area, it is still a very challenging target since there are various types of road scenarios which require a very high robustness of the algorithm. And considering the rather high speed of the vehicles, high efficiency is also a very important requirement for practicable application of autonomous driving. In this paper, we propose a deep convolution neural network based lane detection method, which consider the lane detection task as a pixel level segmentation of the lane markings. We also propose an automatic training data generating method, which can significantly reduce the effort of the training phase. Experiment proves that our method can achieve high accuracy for various road scenes in real-time.
ER -
English
