Road extraction from remote sensing data has been of considerable interest
in recent years due to the rapid progress of geographic information systems
(GIS) and the increasing importance of roads in our daily life. Detailed and
up-to-date road information is an important issue for numerous applications.
Logistics, tourism, car navigation systems are just a few fields of interest. Yet, to accommodate for the needs of these applications, digital road information requires frequent updates, whereby the main source for road data collection is digital aerial and satellite imagery. Despite of numerous
technological advances, the process of data acquisition still needs lot of
manual interaction of a human operator, which is of course both timeconsuming and expensive. Consequently, much effort has been put into
automatic road extraction approaches in recent years (Stilla et al., 2005).
Besides well-known drawbacks due to the specific viewing geometry and
coherent imaging, synthetic aperture radar (SAR) holds some prominent
advantages over optical images (Stilla and Soergel, 2006). For instance, SAR
is an active system, which can operate during day and night. It is also nearly
weather-independent and, moreover, during bad weather conditions, SAR
is the only operational system available today. Road extraction from SAR
images therefore offers a suitable complement or alternative to road extraction from optical images.
The remainder of this chapter is organized as follows: Section 10.2
sketches typical approaches for road extraction from remote sensing data
with their underlying models and extraction strategies before outlining an
extraction system developed at Technische Universitaet Muenchen (TUM)
in more detail. Section 10.3 focuses then on the adaptation of this (generic)
approach to the particular challenges induced by the utilization of SAR
images. Finally, Section 10.4 presents novel concepts for the extension of
road extraction from SAR data concerning the employed models and strategies and illustrates the benefits of these extensions by examples.