Abstract
Road detection is an important research subject in autonomous driving. Both accuracy and efficiency are very important for road detection used in autonomous driving systems. However, these two properties are usually contradictory under certain calculation resources. In this paper, we make a good compromise between accuracy and efficiency by proposing a dual-stage detecting strategy, which consists of a fast Hough transform based road detection method and a reliable vanishing point based method. A dynamic region of interest (ROI) is proposed as a connector of the two stages. Experiments prove that our method can achieve good performance on both accuracy and efficiency.
Original language | English |
---|---|
Title of host publication | Proceedings of 2017 9th International Conference on Computer and Automation Engineering, ICCAE 2017 |
Publisher | Association for Computing Machinery |
Pages | 155-162 |
Number of pages | 8 |
Volume | Part F127852 |
ISBN (Electronic) | 9781450348096 |
DOIs | |
Publication status | Published - 2017 Feb 18 |
Event | 9th International Conference on Computer and Automation Engineering, ICCAE 2017 - Sydney, Australia Duration: 2017 Feb 18 → 2017 Feb 21 |
Other
Other | 9th International Conference on Computer and Automation Engineering, ICCAE 2017 |
---|---|
Country | Australia |
City | Sydney |
Period | 17/2/18 → 17/2/21 |
Keywords
- Hough transform
- Region of interest
- Road detection
ASJC Scopus subject areas
- Human-Computer Interaction
- Computer Networks and Communications
- Computer Vision and Pattern Recognition
- Software