Incremental road discovery from aerial imagery using curvilinear spanning tree (CST) search

Guozhi Wang; Yuchun Huang; Rongchang Xie; Hongchang Zhang

doi:10.1117/12.2241945

26 October 2016 Incremental road discovery from aerial imagery using curvilinear spanning tree (CST) search

Guozhi Wang, Yuchun Huang, Rongchang Xie, Hongchang Zhang

Proceedings Volume 10008, Remote Sensing Technologies and Applications in Urban Environments; 100080S (2016) https://doi.org/10.1117/12.2241945
Event: SPIE Remote Sensing, 2016, Edinburgh, United Kingdom

Abstract

Robust detection of road network in aerial imagery is a challenging task since roads have different pavement texture, road-side surroundings, as well as grades. Roads of different grade have different curvilinear saliency in the aerial imagery. This paper is motivated to incrementally extract roads and construct the topology of the road network of aerial imagery from the higher-grade-first perspective. Inspired by the spanning tree technique, the proposed method starts from the robust extraction of the most salient road segment(s) of the road network, and incrementally connects segments of less saliency of curvilinear structure until all road segments in the network are extracted. The proposed algorithm includes: curvilinear path-based road morphological enhancement, extraction of road segments, and spanning tree search for the incremental road discovery. It is tested on a diverse set of aerial imagery acquired in the city and inter-city areas. Experimental results show that the proposed curvilinear spanning tree (CST) can detect roads efficiently and construct the topology of the road network effectively. It is promising for the change detection of the road network.

Conference Presentation

Citation Download Citation

Guozhi Wang, Yuchun Huang, Rongchang Xie, and Hongchang Zhang "Incremental road discovery from aerial imagery using curvilinear spanning tree (CST) search", Proc. SPIE 10008, Remote Sensing Technologies and Applications in Urban Environments, 100080S (26 October 2016); https://doi.org/10.1117/12.2241945

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available