B3. Istanhul 2004
ing threshold and
are extracted as a
are used for next
is considered to
converges at 4
an 10 pixels are
metric relation is
of small segments
netric information
e of improvement
vhere photometric
mall segments are
mes 50% though
cess. The original
dge extraction are
linking threshold
0 pixels are not
ing process.
| image have been
s of three stages;
I grouping stage.
ohtness threshold
brighter / darker
method has been
of parameters has
r. It suggests the
rue road segments
' been also tested
is same as that of
and radiometric
orrectness though
to rise both of
to achieve it is
in pre-processing
| as a guide and
ao (2002) use the
grouping to “the
s were added and
d might be worth
ind Revision by
atial Data and
ards Operational
ein, H. W., and
Based on Multi-
pp. 777 — 785.
d for Extracting
'ision and Image
1 Road Extraction
RS, vol. 34, part
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
Park, S., and Kim, T., 2001. Semi-Automatic Road Extraction
Algorithm from IKONOS Images Using Template Matching.
Proceeding of the 22nd ACRS.
Price, K., 1999. Road Grid Extraction and Verification. IAPRS,
vol.32, Part 3-2W5, pp. 101-106.
Steger, C.. 1998. An Unbiased Detector of Curvilinear
Structures. IEEE Trans. PAMI, 20(2), pp.113-125.
Treash, K., and Amaratunga, K., 2000. Automatic Road
Detection in Grayscale Aerial Images. ASCE Journal of
Computing in Civil Engineering, 14(1), pp. 60-69.
Zhao, H., Kumagai, J., Nakagawa, M., and Shibasaki, R., 2002.
Semi-automatic Road Extraction From High-resolution Satellite
Image. Proceedings of Photogrammetric Computer Vision, A,
pp. 406-411.
ACKNOWLEDGEMENT
The author would like to thank Ministry of Education, Culture,
Sports, Science and Technology for provide a scholarship. The
author also greatly appreciates Prof. I. Dowman and staff &
member of the University College London for making
opportunity to study as a honorary research fellow and to have
useful discussions.
