well reserved 
'an-sharpened 
isually better 
igh resolution 
(Jensen et al. 
unsupervised 
classify the 
an-sharpened 
the classified 
eure 3a). It is 
tly extracted. 
For example, 
oad networks, 
jad networks. 
road network 
Istanbul 2004 
  
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
   
  
2.3 Edge Detection 
Sobel, Robert and Canny detectors were compared in this study. 
Robert edge detector can easily achieve a clear and proper edge 
image from a QuickBird Pan image (Figure 4a and 3b). 
However, some detailed edges in indistinct edge areas cannot be 
detected. The Canny edge detection algorithm (Canny, 1986) 
needs to adjust two thresholds and a standard deviation of 
Gaussian smooth mask to yield a proper result. But, edges in 
blurred areas can be clearly delineated. In this study, therefore, a 
combination of Robert and Canny detectors is employed. 
  
Figure 4, Detection of road edges from original Pan image. (a) 
The original QuickBird Pan image. (b) The inverse of binary 
edge image from Robert edge detector. 
e 
2.4 Edge-Aided Classification 
Edge-Aided Segmentation. As shown in Figure 3b, the road 
network classified from the pan-sharpened image contains 
many non-road objects either connecting to or isolating from 
the road network. Currently, most existing road extraction 
methods (e.g., Doucette et al. 2001, Zhang et al. 1999) 
experience difficulties to deal with such problems. In this 
study, therefore, we utilize the edges from the corresponding 
Pan image to separate the non-road objects from the road 
network. After performing the edge-aided segmentation, those 
objects connected to road networks are disconnected from the 
road networks. This can be clearly seen by comparing Figure 
3b and Figure 5. 
% 
ne 
* 
- sm. -— _ = 
Figure 5. Road networks after edge-aided segmentation. 
Shape-based Segmentation. A fast component labelling 
algorithm is applied to the road image after disconnecting 
noise, e.g. drive ways and house roofs, from the classified road 
network. Individual objects, including road networks and noise, 
are labelled first. They are then segmented according to their 
size (number of pixels) and shape information (e.g. 
compactness), resulting in final road networks to be extracted 
(Figure 6). An iterative process of edge-aided segmentation, 
shape-based segmentation, segments filtering, and mathematic 
morphological operations may be needed to deal with complex 
cases. 
An edge-aided classification approach was developed to extract 
   
   
        
ve accurate road networks from a classified road image with the 
of the help of the edge information from the corresponding Pan image. 
Figure 6. Road network extracted after edge-aided 
classification process. 
:d road image Tha ; À : E e ; 
ed 5 The edge-aided classification consists of three main processes: 
edge-aided segmentation, shape-based segmentation, and 
segments filtering.