General Road Detection from a Single Image
Automated vehicle navigation systems need to be able to utilize image-based road detection algorithms to be successful. This article consider the problems that a computer may have in identifying roads that may not be well-paved, or have clearly delineated edges, or some known color or texture distribution. The authors separate the road detection process into two steps: the estimation of the vanishing point associated with the main (straight) part of the road, followed by the segmentation of the corresponding road area based upon the detected vanishing point. Their proposed approach includes a novel adaptive soft voting scheme based upon a local voting region using high-confidence voters, whose texture orientations are computed using Gabor filters, and a new vanishing-point-constrained edge detection technique for detecting road boundaries. They report on actual use with their proposed method, including experiments with 1,003 general road images. Results show that this method is efficient, can be run in real-time, and is effective at detecting road regions in challenging conditions.
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/10577149
-
Authors:
- Kong, Hui
- Audibert, Jean-Yves
- Ponce, Jean
- Publication Date: 2010-8
Language
- English
Media Info
- Media Type: Print
- Features: Figures; Photos; References;
- Pagination: pp 2211-2220
-
Serial:
- IEEE Transactions on Image Processing
- Volume: 19
- Issue Number: 8
- Publisher: Institute of Electrical and Electronics Engineers (IEEE)
- ISSN: 1057-7149
Subject/Index Terms
- TRT Terms: Algorithms; Automated vehicle control; Computer models; Detection and identification systems; Estimating; Image analysis; Proximity detectors; Real time information; Road shoulders; Roadside; Rural highways
- Subject Areas: Data and Information Technology; Highways; I72: Traffic and Transport Planning;
Filing Info
- Accession Number: 01352283
- Record Type: Publication
- Files: TRIS
- Created Date: Sep 21 2011 7:14AM