Fast Lane Detection Based on Deep Convolutional Neural Network and Automatic Training Data Labeling

Xun PAN  Harutoshi OGAI  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E102-A   No.3   pp.566-575
Publication Date: 2019/03/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E102.A.566
Type of Manuscript: PAPER
Category: Image
Keyword: 
real-time lane detection,  deep neural network,  automatic labeling,  

Full Text: PDF(7.1MB)
>>Buy this Article


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.