ul 2004 
  
  
  
  
tion 
ugh 
nds, 
with 
the 
ting 
n of 
87). 
ting 
ices, 
ture 
long 
sing 
N 
\ 
me | 
tation | 
  
was 
med 
| of 
ance 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
   
   
    
ground Truth Raster,.. / GroundTruthl 
= Khat = 
~~ 
Figure 4. Error matrix of classification which performed 
on feature components. 
3. Discussion and Conclusion 
The results obtained from this study show that Principal 
Components of raw bands and vegetation indices can 
extract valuable and concentrated vegetation information 
by creating a new variable set with eliminated interband 
correlation and reduced dimensionality of the data. In this 
method Principal Components which were highly loaded 
with the spectral information of desired band or index 
| considered as a feature component, and used in the 
classification process. By using this method the accuracy 
of classification could be increased up to 15%. 
This is a simple and fast technique which could easily be 
implemented on a landscape scaled classification studies. 
4. Acknowledgements 
We would like to thank WWF-Turkey for providing the 
necessary data for this study within the project “Priority 
| Forest Areas of Mediterranean Turkey”. 
5, References 
Brook, R. K. and Kenkel, N. C., 2002. A multivariate 
approach to vegetation mapping of Manitoba's 
Remote Sensing, vol. 23, no. 21, pp. 4761- 4776. 
Bryant, J., 1988. On displaying multispectral imagery. 
Photogrammetric | Engineering | and Remote 
Sensing, vol. 54, pp.1739-1743. 
| Drake, N. A., Mackin, S., Settle, J. J., 1999. Mapping 
Vegetation, Soils, and Geology in Semiarid 
Shrublands Using spectral Matching and Mixture 
Modeling of SWIR AVIRIS Imagery. Remote 
Sensing of Environment, vol. 68, pp. 12-25. 
Fahsi, A., Tsegaye, T., Tadesse, W., Coleman, 2000. T. 
Incorporation of digital elevation models with 
Landsat-TM data to improve land cover 
classification accuracy. Forest Ecology and 
Management, vol. 128, pp. 57-64. 
Gillespie, A.R., Kahle, A.B. and Walker, R.E., 1987, 
Color enhancement of highly correlated images. 
Decorrelation and HSI contrast stretches. Remote 
Sensing of Environment, vol. 20, pp. 209-235. 
Hudson Bay Lowlands, International Journal of 
207 
Ricotta, C., Avena,G.C., Volpe, F, 1999. The influence of 
principal component analysis on the spatial 
structure of a multispectral dataset, International 
Journal of Remote Sensing, vol. 20, no. 17, pp. 
3367- 3376. 
Sabine, C., 1999, Remote sensing strategies for mineral 
exploration. Remote Sensing for the Earth Sciences 
— Manuel of Remote Sensing, 3rd edn, editted by 
A. Rencz ( New York: American Society of 
Photogrametry and Remote Sensing/ John Wiley 
and Sons), pp.375-447. 
| 
| 
if 
TL 
EARS 
EU E