nbul 2004 International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
  
image and the proportion images of threshed cornfield, 
umination deciduous forests, cornfield and buildings. For example, the 
hood and brighter the pixel of the buildings proportion image is the larger 
ssification the proportion of the class is in the pixel. It means that large 
r and the buildings have brighter pixel. The results for coniferous forest 
and road proportion images were worse. 
4. CONCLUSIONS 
Generally speaking the classification results were good. For 
example, Maximum Likelihood classifier led to good results 
(overall accuracy was about 91 percent), but it requires more 
computation time. Spectral Angle Mapper and Spectral 
Correlation Mapper were faster and they led to better 
classification results in poor illumination. The results of 
Minimum Distance classification were poor. Spectral Unmixin 
algorithm worked in some cases. It produced good proportion 
images for threshed cornfield, deciduous forests, cornfield and 
buildings but coniferous forest and road did not work. The 
suitable reference spectra for the mixed pixel were hard to find. 
There was too much variation in the pixel values of the same 
class. This study shows clearly that it is worthwhile to pay 
attention to different methods when the reference spectra are 
  
calculated. 
5. REFERENCES . 
ts of 
num Boardman, J., 1994. Geometric mixture analysis of imaging 
spectrometry data. In: International Geoscience and Remote 
Sensing Symposium, IGARSS '94, Surface and atmospheric 
ze of the remote sensing: technologies, data analysis and interpretation. 
ginal size California, USA, Vol. IV, pp. 2369-2371. 
d by only 
reference Carvalho, O. & Meneses, P., 2000. Spectral Correlation Mapper 
sification. (SCM): An Improving Spectral Angle Mapper. In: Ninth JPL 
ated from Airborne Earth Science Workshop. Pasadena, CA, JPL 
here pixel Publication 00-18, pp. 65-74. 
Is (Figure 
cent and Kallio, K., Kutser, T., Hannonen, T., Koponen, S., Pulliainen, 
J., Vepsäläinen, J., Pyhälahti, T. 2001. Retrieval of water 
quality from airborne imaging spectrometry of various lake 
types in different seasons. Science of the Total Environment, 
268, pp. 59-77. 
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. 
  
Mäkisara, K. &  Tomppo E. 1996. Airborne imaging 
spectrometer in National Forest Inventory. In: International 
Geoscience and Remote Sensing Symposium, IGARSS '96, 
Remote sensing for a sustainable future. New York, USA, Vol. 
II, pp. 1010-1013. 
Ruohomäki, T., Tórmà, M., Rainio, H., Lumme, J. & Mäkilä, 
M. 2002. Soil types, vegetation and AISA-imagery in Lammi 
test area. Geological Survey of Finland, GTK/RS/2002/6 
  
ixing 
lal image 
e original 
87