The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B3b. Beijing 2008
Fig. 5 is another example of multi-spectral IKONOS image of
Lavasan (4m resolution and 169x162 pixels) in semi-urban
region. In this figure square object that is the place of
intersection of three roads, is shown. Again the described
procedure of road extraction is implemented on the image. As
you see in these figures, the cause of producing gaps on the
roads is existence of obstacles such as trees, curbs and cars .
these gaps are filled using morphological operators.
The most important problem in fig. 5 is limitation of MST
algorithm in closing loops. This restriction causes missed links
in extraction of cyclic objects like squares which are shown in
fig. 5g by yellow lines. This restriction leads the extraction
process to become semi-automatic and supervision of human
operator for complement of missed links is compulsory.
Jfi
Si
&
| Grid dim.= 8x12
| Active Node no.=33 |
'-O
II
§
£
1
Q
X
II
i
nO
rsa
(h
o
i
i
z
1
2c
\o
IT
s
z
1
Q
Grid dim. = 13x12 |
Acth'eNode no.=71
<»
II
i
£
1
Q
E
S
g S
u. 2
£
ii
W
£
•a
TJ
1
3
w
T
£
-»J
.a
1
3
w
CO
II
nr ;
£
-a
no
1
3
w
1
is
c
&
-
1 &
r. Z
CQ
-
■ef
Tt
Of
j§
if
o a
1-1 &•
in
zf “
m rf,
z ® ^
6
z
Uh
>■
<
<N
C3
H
CONCLUSIONS
In this paper a semi-automatic road extraction system based on
combination of different feature extraction techniques
containing FCM and C-Means clustering technique,
morphological functions and graph theory is proposed.
Input images of this system consist of multi-spectral and pan-
sharpened IKONOS images of Lavasan city in Iran (with
respectively 4 and 1 meters spatial resolution).
The main advantage of this proposed system is achievement of
it in extracting different shaped roads such as straight, spiral,
junction and square and attaining acceptable precisions in order
to updating road maps. The only drawback of this system is
limitation in completely extraction of road center line in place
of squares and closed loops. So supervision of human operator
for completing missed links and closing the loops is inevitable.
Attaining mean overall accuracy (OA) of 98.2% and Kappa
coefficient of 86.26% in classification of image to road and
non-road classes, and also mean RMS error of 0.64 pixel in
comparing automatic extracted road centerline with manual
extracted one, are a good criterion of proposed system success
in semi-automatic extraction of road.
REFERENCES
Agouris, P., Doucette, p., Stefanidis, A., Spatiospectral
Cluster Analysis of Elongated Regions in Aerial Imagery, IEEE
International Conference on Image Processing (ICIP) ,
Thessaloniki, Greece, Vol. ^, pp. VAt.vtr.
Doucette, P., Agouris, P., Stefanidis, A., Musavi, M., 8 • • L
Self-organised clustering for road extraction in classified
imagery. ISPRS J. Photogr. Remote Sensing, v 00 . Viv.roA.
Ebner, H., Eckstein, W., Heipke, C., Mayer, H., ^5 5.
Automatic extraction of GISobjects from digital imagery.
Intemat. Arch. Photogrammet. Remote sensing
F. Hoppner , Fuzzy Shell Clustering Algorithms in Image
Processing: Fuzzy C-Rectangular and 2-Rectangular Shells,
IEEE Trans, on Fuzzy Systems 5 (1997) 599-613.
Gonzalez, R.C., Woods, R.E.,. Digital image processing,
Prentice-Hall, Inc, Second edition, * • • V
Horowitz, E., Sahni, S., ^5Ar. Fundamentals of Data Structures,
GALGOTIABOOKSOURCE.
http://en.wikipedia.org/wiki/Data_clustering
HUNG, M., WU+, J„ CHANG, J„ YANG, D„ 2005. An
Efficient k-Means Clustering Algorithm Using Simple
Partitioning , JOURNAL OF INFORMATION SCIENCE
AND GINEERING 21, 1157-1177
Liu, H., Li, J, and Chapman, M. A., * • Automated Road
Extraction from Satellite Imagery Using Hybrid Genetic
Algorithms and Cluster Analysis, Journal of Environmental
Informatics, Vol A, No A, pp. * *-*v
Modenesi, M., Costa, M., Evsukoff, A., Ebecken, N.,
Parallel Fuzzy c-Means Cluster Analysis. VECPAR, v th
international meeting, Brazil.
Mohammadzadeh, A., Tavakoli, A., Valadanzoej, M., ^..1.
Road extraction basedon fuzzy logic and mathematical
morphology from pan-sharpened ikonos
images,Photogrammetric Record, I'('A); i £-1 •
R. N. Dave, 1989. Use of the Adaptive Fuzzy Clustering
Algorithm to Detect lines in Digital Images, Intell. Robots
Comput. Vision VIII 1192 - 600-611.