International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
Accuracy of pixel- and standwise tree species classification 
100 1 
  
30 
Classification accuacy (76) 
3 4 
Featureset 
Figure 4. The classification accuracies of pixelwise and 
standwise tree species classifications. The solid line with "o" 
represents the overall accuracies, solid line the mean of user's 
and producer's accuracies of pine, dashed spruce and dashdot 
deciduous trees. The accuracies of standwise classifications 
have been represented with "x". 
5. CONCLUSIONS 
These results illustrate that it is difficult to make accurate land 
cover classification using one frequency, one polarization 
microwave information even if there are some temporal 
resolution or low resolution optical data available. More 
information about the target would be needed, like 
interferometric coherence, or more than one frequency or 
polarization. The feature extraction increased classification 
accuracy a bit but not much. The standwise classification 
increases accuracy due to feature averaging. 
The decision based data fusion using a'posteriori probabilities 
of low resolution classification as a'priori probabilities of high 
resolution classification shows promise, but needs some 
development. The increase of overall and classwise accuracies 
in this study were more than 10 and 25 %-units, respectively. 
One alternative could be to compute lower resolution images 
from ERS-images and form them to hierachical series. 
Interpretation would begin from lowest level and higher level 
could use lower-level interpretation as input. 
ACKNOWLEDGEMENTS 
The authors would like to thank National Technology Agency 
Tekes for financial help. European Space Agency has provided 
ERS SAR-images under AO-project "Retrieval of boreal forest 
and surface characteristics from Envisat multisensor data". 
REFERENCES 
Devivjer P., Kittler J., 1982. Pattern Recognition - A Statistical 
Approach. Prentice-Hall, 1982. 
Gaveau, D., Balzter, H., Plummer, S., 2003. Forest woody 
biomass classification with satellite-based radar coherence over 
900 000 km2 in Central Siberia. Forest Ecology and 
Management, 174(1-3) , pp 65-75. 
Haralick, R., Shanmugam, K., Dinstein, L, 1973. Textural 
Features for Image Classification. [EEE Transactions on 
Systems, Man. and Cybernetics, 3, pp. 610-621. 
Herold. N.. Haack, B., Solomon, E., 2004. An evaluation of 
radar texture for land use/cover extraction in varied landscapes. 
International Journal of Applied Earth Observation and 
Geoinformation, 5(2), pp. 113-128. 
Kramer, H., 1996. Observation of the Earth and Its 
Environment. Springer, 1996. 
Kruse, F., Lefkoff, A., Boardman, J., Heidebrecht, K., Shapiro, 
A.. Barloon, P., Goetz, A. 1993. The spectral image processing 
system (SIPS) - interactive visualization and analysis of 
imaging spectrometer data. Remote Sensing of Environment, 
44(2-3), pp. 145-163. 
Kruse, F.. Richardson, L. & Ambrosia, V. 1997. Techniques 
developed for geologic analysis of hyperspectral data applied to 
near-shore hyperspectral ocean data. Fourth International 
Conference on Remote Sensing for Marine and Coastal 
AEnvironments. Orlando, Florida, Vol. I, pp. 233-246. 
Kurvonen, L., Pulliainen, J., Hallikainen, M., 2002. Active and 
passive microwave remote sensing of boreal forests. Acta 
Astronautica, 51(10), pp. 707-713. 
Lillesand T., Kiefer R., 1994. Remote Sensing and Image 
Interpretation. John Wiley & Sons, Inc, 1994. 
Masuoka, E. Fleig, A. Wolfe, R., Patt, F, 1998. Key 
Characteristics of MODIS Data Products. /EEE Transactions on 
Geoscience and Remote Sensing, 36(4), pp. 1313 — 1323. 
Pohl, C., van Genderen, J., 1998. Multisensor Image Fusion in 
Remote Sensing: Concepts, Methods and Applications. 
International Journal of Remote Sensing, 19(5), pp. 823-854. 
Ranson, K, Sun, G., Kharuk, V. Kovacs K. 2001. 
Characterization of Forests in Western Sayani Mountains, 
Siberia from SIR-C SAR Data. Remote Sensing of Environment, 
75(2), pp. 188-200. 
Sui s yos s ond 
Richards, J., 1993. Remote Sensing Digital Image Analysis. 2" 
ed., Springer, 1993. 
Schneider, A., Friedl, M., Mclver, D., Woodcock, C., 2003. 
Mapping Urban Areas by Fusing Multiple Sources of Coarse 
Resolution Remote ly Sensed Data. Photogrammetric 
Engineering and Remote Sensing, 69(12), pp. 1377-1386. 
Tomppo E., 1991. Satellite Image Based National Forest 
Inventory of Finland. International Archives of 
Photogrammetryand Remote Sensing. 28: 419-424. 
Tórmá, M., 1999. Classification of Tree Species Using ERS 
Intensity and Coherence Images. Second International 
Workshop on ERS SAR Interferometry, 10.-12.11.1999 Liege 
Wang, F., 1990. Fuzzy supervised classification of remote 
sensing images. IEEE Transsactions on Geoscience and Remote 
Sensing, 28, pp. 194-201. 
Wegmiiller, U., Wemer, C. Strozzi, T., 1998. SAR 
Interferometric and Differential Interferometric Processing 
Chain Proceedings of IGARSS'98, Seattle, USA, 6-10 July 
1998, vol.2, pp.1106-1108 
No 
O2 
N 
Eu 
The 
50% 
man: 
infor 
supp 
Lanc 
Furtl 
logg 
Lanc 
sub-j 
and | 
achic 
illeg: 
fellir 
Fore: 
resot 
envir 
of th 
envi 
impo 
as \ 
Defo 
more 
Ana 
the | 
conti 
Serio 
resul 
warn 
has a 
the 
Conf 
in Ju 
this r 
Indor 
total 
10% 
an i 
loggi 
most 
Cutti 
Syste 
activi