International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
shown in Figure 3b. Secondly, both classifiers had difficulties 
to learn the characteristics of road class as it largely included 
mixed pixels due to the pixel resolution of the images (i.e. 
30m). Thirdly, it is found that the addition of first three 
principal components to the classification process did not make 
any improvement. On the contrary, it reduced the classification 
accuracy probably because of the increased complexity and 
dimensionality of the data. This point obviously needs to be 
clarified with further research. Lastly, the classification 
methods, especially ML classifier, show sensitivity to classes 
depending on their spectral variability ML algorithm 
dominated road class over the image whilst ANN classifier was 
slightly sensitive to inland water class. 
ACKNOWLEDGEMENTS 
The authors gratefully acknowledge the financial support from 
Leica-Sistem. 
REFERENCES 
Ashutosh, S., 2002. Principal component-based algorithm on 
multispectral remote sensing data for spectral discrimination of 
tree cover from other vegetation types. Current Science, 82(1), 
pp. 67-69. 
Civco, D. L., 1993. Artificial neural networks for land-cover 
classification and mapping. International Journal of 
Geographical Information Systems, 7(2), pp. 173-186. 
Garson, G. D., 1998. Neural Networks: An Introductory Guide 
for Social Scientists. SAGE Publications, London. 
Gopal, S. and Woodcock, C., 1996. Remote sensing of forest 
change using artificial neural networks. /EEE Transactions on 
Geoscience and Remote Sensing, 34(2), pp. 398-403. 
Kavzoglu, T. and Mather, P.M., 1999, Pruning artificial neural 
networks: an example using land cover classification of multi- 
sensor images. International Journal -of Remote Sensing, 
20(14), pp. 2787-2803. 
Kavzoglu, T. and Mather, P.M., 2003. The use of 
backpropagating artificial neural networks in land cover 
classification. International Journal of Remote Sensing, 
24(23), pp. 4907-4938. 
Mather, P. M., 1999. Computer Processing of Remotely-Sensed 
Images: An Introduction, Second Edition, John Wiley & Sons, 
Chichester. 
Monger, M., 2002. Soil classification in arid lands with 
Thematic Mapper data. Terra, 20(2), pp. 89-100. 
Paola, J. D. and Schowengerdt, R. A., 1995. A detailed 
comparison of backpropagation neural network and maximum 
likelihood classifiers for urban land use classification. /EEE 
Transactions on Geoscience and Remote Sensing, 33(4), pp. 
981-996. 
Sunar, F., 1998. An analysis of changes in a multidate dat set: 
a case study in the Ikitelli area, Istanbul, Turkey. International 
Journal of Remote Sensing, 19(2), pp. 225-235. 
Sunar Erbek, F., Ozkan, C. and Taberner. M. 2004. 
Comparison of maximum likelihood classification method with 
supervised artificial neural network algorithms for land use 
activities. International Journal of Remote Sensing, 25(9), pp. 
1733-1748. 
956 
W 
lo 
dif 
wi 
m 
int 
pr 
thi 
th 
pa 
dit 
19 
WI 
pro 
im: 
tha 
enc 
wil 
Bu 
hig 
low 
wit 
con 
tim 
req 
For 
its : 
Def 
oft 
Mu 
dat: 
datz