anbul 2004
eature
a from the
using tonal
he correct
1e. But the
wk tone. It
ly detected
1 also be
in feature
eature
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
Mo hu ans
Ee e ee ri
au
PES
b. Using contrast feature c. Using Dissimilarity feature
Figure 5. Extraction of residential area texture feature
Figure 4 b and figure 5 b, c respectively represent the
residential area from the below texture images. We can use the
same texture analysis methods to extract the edge of residential
area feature.
43 Image Interpretation
Different surface features exhibit different scattering
characteristics. From the view of tone, urban areas put up very
strong backscatter; forest puts up medium backscatter; calm
water puts up smooth surface and low backscatter; rough sca
puts up increased backscatter due to wind and current effects.
In the above figure, we can see some mapping elements through
the testing AIRSAR data, such as road, water area and
residential area.
Figure 6. Mapping element extraction
In figure 6, we make our efforts to take mapping element
extraction. Results show that extracting basic mapping element
is feasible.
163
5. CONCLUSIONS
The purpose of this study is twofold: first, to demonstrate the
advantages of using AIRSAR data for topographic mapping
purposes. Second, to demonstrate the advantages of utilizing the
high resolution (0.5-meter) AIRSAR sensor data acquired for
surface texture analysis and interpretation purposes. Finally,
application of SAR images gives satisfactory results from above
experiments.
At present, with the speedy development of China, the timely
repairing and updating map, establishing periodically updating
geography databases, dynamic monitoring land use change
conditions, and deriving various kinds of latest thematic map
are the imperative problems. The first important factor that
restricts this kind of dynamic monitoring is whether we can
provide the practicable, high-resolution, continuously stable and
rapidly receiving and useful data sources or not. [t is shown that
AIRSAR with its full-time and all-weather characteristic
becomes optimum remote sensing data sources solving the
tradition difficulty district in the topography of Surveying and
Mapping.
The various methods for modelling textures and extracting
texture features can be applied in four broad categories of
problems: texture segmentation, texture classification, texture
synthesis, and shape from texture. From the above study, we
could perform texture classification through identifying some
types of homogeneous regions, and texture segmentation
through finding the texture boundaries.
6. ACKNOWLEDGMENTS
The testing AIRSAR image data are provided by 38th institute
and CATIC SIWEL
7. REFERENCES
Du, L. I, 1990. Texture Segmentation of SAR Images Using
Localized Spatial Filtering, [International Geoscience and
Remote Sensing Symposium, 4, pp. 1983-1986.
Jiang, Q. X., Liu H. P., KONG L. Y., 2003. The Application of
Texture Analysis in TM Image Information Extraction, Remote
Sensing Information, pp. 1983-1986.
Lee, J. H. and W. D. Philpot, 1990. A Spectral- Textural
Classifier for Digital Imagery, International Geoscience and
Remote Sensing Symposium, pp. 2005-2008.
Mihran, T., 1998. The Handbook of Pattern Recognition and
Computer Vision, pp. 207-248.
Rignot, E. and Kwok, R., 1990. Extraction of Textural Features
in SAR Images: Statistical Model and Sensitivity, /nternational
Geoscience and Remote Sensing Symposium, pp. 1979-1982.
Yang X. M., Zhou C. H., 1998. Report on ACRS "Recognition
of Flooded Arca in Radar Image using Texture Feature
Analysis”, Manila.
http://www .gisdevelopment.netaars/acrs/1998/ps1/ps1022.shtm
| (accessed 6 May. 2004)