Share Email Print

Proceedings Paper

Host lane detection method using semantic segmentation combined with hierarchy clustering algorithm
Author(s): Yikang Gao; Haiying Wang
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Lane detection is the core module of autonomous cars and plays an important role in lane keeping and trajectory planning. Traditional works on lane detection depend on hand-crafted features and are weak to road scene variations. Recent end-to-end approaches leverage deep learning models, but are computationally demanding and need heavy work of labeling lanes. In this paper, we combine the advantage of traditional method and deep learning based method. Our lane detection method has three key novelties: (1) we use semantic segmentation based deep learning to extract region of interest (ROI), which can free the work of manually selecting features and fast locate the lane area. (2) We propose a lane feature points extracting algorithm based hierarchical clustering to effectively remove the disturbance of the noise. (3) we make use of the similarity of the inter-frame to correct the lane fitting results. Experimental results show that our lane detection method can adapt to various road scene and significantly decrease the false positive rate.

Paper Details

Date Published: 6 May 2019
PDF: 8 pages
Proc. SPIE 11069, Tenth International Conference on Graphics and Image Processing (ICGIP 2018), 110692H (6 May 2019); doi: 10.1117/12.2524180
Show Author Affiliations
Yikang Gao, Beijing Univ. of Posts and Telecommunications (China)
Haiying Wang, Beijing Univ. of Posts and Telecommunications (China)

Published in SPIE Proceedings Vol. 11069:
Tenth International Conference on Graphics and Image Processing (ICGIP 2018)
Chunming Li; Hui Yu; Zhigeng Pan; Yifei Pu, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?