A reliable and fast ribbon road detector using profile analysis and model‐based verification

$61.74 plus tax (Refund Policy)

Buy Article:

Abstract:

Ribbon roads are typical and important geospatial features. In this paper, we proposed a model-based method for extracting ribbon road features from remotely sensed imagery. Firstly, by analysing the perpendicular profiles along the road direction, we use binary profile template matching along the crossing-directions to obtain the coarse candidate road centre points. After tracing the curve segments and analysing the perpendicular profiles, the width of the road ribbons and their lateral sides and the accurate centrelines are located. Secondly, a quantitative ribbon road model, which integrates the geometry and radiometry characteristics, is deployed to verify each extracted road segment. The coarse-to-fine method makes use of an explicit road profile model and overcomes the negative influences of asymmetrical lateral contrast and width variation. Finally, the model-based verification enables more reliable sequential processing, such as perceptual grouping. We have conducted extensive experiments on verifying the algorithm. It has been demonstrated that the developed method is highly reliable for automatic detection of the typical ribbon road features from imagery.

Document Type: Research Article

DOI: http://dx.doi.org/10.1080/0143116042000298243

Affiliations: Geospatial Information and Communication Laboratory, Department of Earth & Space Science and Engineering, York University, 4700 Keele Street, Toronto, ON, Canada, M3J 1P3, {xyhu, tao}@yorku.ca

Publication date: March 1, 2005

More about this publication?
Related content

Share Content

Access Key

Free Content
Free content
New Content
New content
Open Access Content
Open access content
Subscribed Content
Subscribed content
Free Trial Content
Free trial content
Cookie Policy
X
Cookie Policy
ingentaconnect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more