rd 
N 
Tect 
with 
the 
ctor 
rom 
the 
tate 
ture 
the 
)hic 
ure 
al., 
da 
e to 
'ess 
| of 
nce 
and 
the 
pad 
ion 
ds. 
tics 
ied 
WS 
ly 
ige 
ad 
Ice 
the 
Auclair Fortier, Marie-Flavie 
  
snake approach. The correction part can be summarized by these five steps: 1) resampling the image 2) extraction of lines 
3) extraction of line junctions 4) displacement of road segments according to the difference between their extremities and 
image junctions matched, and 5) correction of road segment location with snake process. 
The second step in updating road maps is the addition of new roads. We know that roads are built in networks and that 
new roads are linked with old ones. We thus generate hypothesis for new roads by following lines from line junctions 
near the known network. 
3 DETECTION LINES AND OF LINE JUNCTIONS FROM THE IMAGE 
In Aerial and Satellite images, roads are long, thin elements. Usually, they are bright lines against a dark background. 
To detect these, we use Steger's line detector (Steger, 1998) which can be summarized as follows: A bright line is a 
maximum in the gray-level image in a particular direction. Thus, in this direction, the second directional derivative has 
a negative minimum and its first directional derivative is 0. This detector provides a line plausibility (Amaz (x, y)) and a 
line direction (7(x, y)) function. 
In the context of road detection, road intersections are represented in aerial images by line junctions. Image junctions are 
very useful features, particularly in solving matching problems, because they are generally reliable and stable elements. 
Unfortunately, we have noticed that they are not used for road detection. One reason is that, to the best of our knowledge, 
before the beginning of this project, no line-junction detector was existing. We use a line-junction detector developed in 
our Computer Vision Laboratory, based on a measure of curvature between the direction vectors of the line pixels within 
a given neighborhood (Deschênes and Ziou, 1999). This detector also finds line terminations. This measure is estimated 
by computing the mean of dissimilarity between the orientation vector of the line pixels: 
M M 
1 
+ X Y BAzA)[- psy) da + Ay + Ay) (1) 
Az=—M Ay=—M 
where zi(z, y) is the normalized line direction at point (z, y), - is the dot product, 
B(Az, Ay) = eg (Av Ay) ifa line exists between (x, y) and (x + Ax,y + Ay) Q 
0 otherwise 
and N the number of points (x + Ax, y + Ay) such that f Z 0. The junction is localized in a high curvature region using 
a non-maxima suppression. For more details, the interested reader can see the technical report on this detector (Deschénes 
and Ziou, 1999). Figure 1a presents the results of the line-junction detector. 
4 CORRECTION OF ROAD LOCATION 
The correction of road location involves two steps. The first is re-localization of the known roads to bring them closer to 
the final snake's solution. The second step concerns the snake process itself. These two steps are presented in the next 
two sections. 
4.1 Matching the Road Database with Image 
The active contour model (snake) is an iterative process which needs an initialization close to the final solution. This 
initialization must fall within the attraction zone of the snake. We propose to use line junctions to re-localize road location 
from the road database, because they indicate the position of road intersections in the image. For this purpose, we use 
a correlation measure between certain points in the road database and line junctions. The road database provided by 
the Center for Topographic Information of Geomatics Canada is made up of lists of georeferenced points defining line 
segments. We call these line segments vectors. These points are in georeferenced coordinates, thus we have to convert 
them in coordinates relative to the beginning of the image. Then, we group the vectors into road segments in order 
to reduce the number of snakes. These road segments must be as straight as possible because this allows us to add a 
constraint of high rigidity to the snake. This high rigidity will help avoid noise and shadowing problems, because the 
optimized contour does not follow small intensity variations caused by these phenomenons. Two vectors are linked if they 
are adjacent, if the cosine of the angle between them is below a threshold (in our experiments, we always set this threshold 
to cos 138?) and if there is no other vector starting at the extremity common to the two vectors. In other words, we find 
road segment extremities 1) where there are more than two vectors ending at a single point 2) where the angle between 
two vectors is lower than a given value 3) at terminations. A road segment is described by all vectors between its two 
extremities. Figure 1b shows an example of road segments from the database. There are five road segments in this image. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 91