International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
4. ACCURACY ASSESSMENTS
Road completeness and correctness (Heipke et al., 1997) are
used to assess the accuracy of the road extraction. The
completeness is the ratio of correctly extracted road length
(length matched between the extracted and reference data) to the
total road length from the reference image. The highest value is
|. The correctness is the ratio of correctly extracted road length
to the total length of the extracted road network. The optimal
value is also 1. In this study, the completeness value reaches
0.94, and the corresponding correctness value is 0.98.
The results demonstrate that the proposed method achieves
significantly higher accuracy than those of Pan based feature
extraction (e.g. Hinz et al., 2001), multi-spectral classification
(e.g. Shackelford et al., 2003), and MS and Pan integrated
classification (e.g. Granzow, 2001).
5. CONCLUSION
A new approach for object extraction. from high-resolution
satellite images has been developed which effectively integrates
image fusion, multi-spectral classification, feature extraction
and feature segmentation into the object extraction process.
Both spectral information from MS images and spatial
information from Pan images are utilized for the extraction to
improve the extraction accuracy. Experiments in road extraction
from QuichBird MS and Pan images demonstrate that the
proposed approach is effective. The completeness and
correctness of road network extraction reaches 0.95,
significantly higher than those of other existing road extraction
methods.
6. ACKNOWLEDGEMENTS
This research is funded by GEOIDE Phase ll, a research
funding program of Canadian Networks of Centres of
Excellence.
REFERENCES
Canny, J.F., 1986. A Computational Approach to Edge
Detection. [EEE Transactions on Pattern Analysis and
Machine Intelligence, 8(6):679-698.
Csatho B.M., and Schenk, A. F., 1998. Multi-sensor Data
Fusion for Automatic Scene Interpretation, /nfernational
Archives of Photogrammetry and Remote Sensing, XXXII
(7), pp. 336-341, Budapest.
Doucette, P., Agouris, P., Stefanidis, A., Musavi, M., 2001.
Self-organised Clustering for Road Extraction in
Classified Imagery. ISPRS Journal of Photogrammetry &.
Remote Sensing 55, 347—358.
Granzow, Ed., 2001, Automatic Feature Recognition and
Extraction from Remote Sensing Imagery. National
Consortia on Remote Sensing in Transportation. US
Department of Transportation.
Heipke, C., Mayer, H., Wiedemann, C., and Jamet, O., 1997.
Evaluation of Automatic Road Extraction, /nternational
Archives of Photogrammetry and Remote Sensing, Vol. 32,
No.3, pp.47-56.
Hinz, S., Baumgartner, A., Mayer, H., Wiedemann, C., Ebner,
H., 2001. Road Extraction Focussing on Urban Areas.
Automatic extraction of Man-Made Objects from Aerial
and Space Images (III) (E.P. Baltsavias, A. Gruen,
L.V.Gool, editors), 4. A. Balkema Publishers, Lisse, pp.
255-266.
Jensen, J.R., Cowen, D.C., Halls, J., Narumalani, S., Schmidt,
N., Davis, B.A., and Burgess, B., 1994. Improved Urban
Infrastructure Mapping and Forecasting for BellSouth
Using Remote Sensing and GIS Technology,
Photogrammetric Engineering & Remote Sensing, 60:339-
346
Zhang, C., and Baltsavias, E.P., 1996. Road Network Detection
by Mathematical Morphology. Proceeding of ISPRS
Workshop: 3D Geospatial Data Production: Meeting
Application | Requirement. 7-9. April. Paris. France.
Pp.185-200.
Zhang, Y., 2002. A New Automatic Approach for Effectively
Fusing Landsat 7 images and IKONOS Images.
c
IEEE/IGARSS '02, Toronto, Canada, June 24-28, 2002.
KEY
ABS
Three
plann
data €
assoc
. by gr
like |
anda
withi
frame
map
in ho
; 1
Thre:
tant
urbai
touri
ing r
proci
(Bail
metr
more
essai
The
ther
imag
For
map
ular
not |
pres
for |
limi
Coar
