chapter  11
22 Pages

Road Networks Derived from High Spatial Resolution Satellite Remote Sensing Data

WithRenaud Pe´teri and Thierry Ranchin

CONTENTS 11.1 Introduction ............................................................................................. 216 11.2 Automatic Extraction of Road Networks: From Linear

to Surface Methods................................................................................. 217 11.2.1 Linear and Surface Representation of Roads ....................... 217 11.2.2 New Sensors, New Cartographic Scale................................. 217 11.2.3 Road Extraction: Linear Approaches..................................... 218 11.2.4 Road Extraction: Surface Approaches................................... 219

11.2.4.1 Introduction .............................................................. 219 11.2.4.2 A Better Precision, New Kinds of Artifacts......... 219

11.3 A Collaborative Method for Surface Road Extraction from High Spatial Resolution Images ................................................. 220 11.3.1 Description................................................................................. 220 11.3.2 Graph Management.................................................................. 221

11.3.2.1 Extraction of the Graph Polylines......................... 221 11.3.2.2 Getting the Complete Graph by Connecting

the Extracted Polylines ........................................... 222 11.3.3 Reconstruction Module............................................................ 222

11.3.3.1 Description................................................................ 222 11.3.3.2 Extraction of Parallel Road Sides .......................... 223 11.3.3.3 Extraction of Road Intersections ........................... 224 11.3.3.4 Reconstruction Algorithm ...................................... 224

11.3.4 Properties of This Approach................................................... 225 11.3.5 Assessment of the Results ....................................................... 225

11.4 Application: QuickBird Image of the Area of Fredericton, Canada........................................................................... 226 11.4.1 Description of the Studied Scene ........................................... 226 11.4.2 Extraction of the Road Network Graph................................ 226 11.4.3 Reconstruction Module............................................................ 228

11.4.4 Quantitative Assessment ......................................................... 229 11.4.4.1 Gain in Time ............................................................. 232

11.5 Conclusion and Prospects ..................................................................... 232 Acknowledgments ............................................................................................. 233 References ........................................................................................................... 233

There is a strong demand for accurate and up-to-date road network information. Road network knowledge is crucial for the creation and the update of maps, geographic information system (GIS) database, transportation, or land planning. For local authorities, cartography of the road network is needed for urban planning, dirty water collection through gutter network (most often located under roads), traffic flow analysis, or pollution mapping. Closely related applications are geomarketing, electricity and telecommunication networks, databases for car navigation, and so on. Currently, road network cartography is essentially done by human interpretations of high-resolution aerial images and additional in situ information. This is a long and tedious work that requires to be done again for each update of the road network.