References 
1. Valerie. L. Thomas, Generation and Physical Characterristics of » 
the Landsat 1 and 2 MSS Computer Compatible Tapes, Goddard Spa- 
“ -cen Flight Center. : 
2. Patrick F. Holkenbrink, Manual on Characteristics of Landsat 
Computer-Compatible Tapes Produced by the EROS Data Center 
Digital Image Processing System, United States Geological Sur- 
vey 1978 
5. Dieter Steiner and Michael E. Kirby, Geometrical Referencing of 
Landsat Images by Affine Transformation and Overlaying of Map 
Data, Photogrammetria 33 (1977) 41-75. 
4, Francis J. Ahern and Jennifer Murphy, Radiometric Calibration 
and Correction of Landsat 1, 2, and 3 MSS Data, Canada Center 
for Remote Sensing Energy Mines and Resources, Ottawa. 
5, K. S. Fu, Digital Pattern Recognition, Springer-verlay Berlin 
Heicklberg New York, 1980. 
6. David Landgrebe, Computer-Based Remote Sensing Technology-A Look 
to the Future, Hemote Sensing of Environment 5,229-246 (1976). 
7. Jack Bryant, On the Clustering of Multidimensional Pictorial 
Data, Pattern Recognition Vol.11 PP 115-125. 
8. Z. Kalensky, ERTS Thematic Map from Multidate Digital Images, 
Prcceeding Symposium on Remote Sensing and Photo Interpretation 
Vol.2. 
9. L. Borriello and F. Capozza, A Clustering Algoritnm for Unsuper- 
vised Crop Classification, CCATA, Bari, Italy. 
10. Torleiv Orhaug and S Ingvar Akersten, Digital Processing of 
Multispectral Data, FOA Repport C30075-E1. 
11. Beguin H. Do Tu H, and Wilmet. J, Comparison of Classification: 
Methods for Urban Images Interpretation,XIV Congress of the 
International Society of Photogrammetry, Commission VII. 
12. Torleiv Orhaug, Leif Wastenson and S Ingvar Akersten, Forest 
Inventory Using Landsat Digital Imagery, FOA Repport A30008-E1. 
13, We A, Davis and F. G. Pett, The Identification and Reclassifica-. 
tion of Small Regions on Digital Thematic Maps, Forest Manage- 
ment Institute Information Report FMR-X-90. 
^4. Z. Kalensky and L. R. Scherlr, Accuracy of Forest Mapping from 
Landsat Computer Compatible Tapes. 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
   
   
  
  
  
  
  
  
    
Les mé 
détection sc 
considèrent 
le contenue 
Dans c 
révèle être 
ces méthodes 
L'inty 
=. à 
difficultés 
a 
classificat! 
Class” 
mainly mult 
as isolated. 
does not ap| 
When > 
results give 
ted. 
Intro« 
mitting to ! 
also yielde 
tion.