100 
The cited drawbacks are also paths towards the possible 
development of the method. In continuation of the forest 
mapping project the improved forest border map will 
enable a more focused classification of the forest types, 
based on the deciduous / broadleaved ratio, age and 
canopy cover. We expect the forest border map of 
Slovenia - when completed - to find applications also 
outside the forestry information system. Since the 
presented method enables a reasonably precise, country 
wide and up-to-date identification of the true forest cover 
as opposed to the often outdated cadastral situation, one 
of the first applications would be to determine the actual 
forest area and map the loss of farmland in the last 
decades due to spontaneous afforestation. Additionally, 
the presented method could provide an inexpensive 
means to monitor the forest cover changes, giving the 
policy makers in physical and social planning a useful tool 
to verify past decisions and plan efficient new actions. 
ACKNOWLEDGEMENTS 
This study is part of the project "Development of 
Multifunctional Forestry Information System", 
financed by the Ministry of Agriculture, Forestry, and 
Food of Slovenia and by the Ministry of Science and 
Technology of Slovenia. We are also grateful to Gal 
Kusar, who photointerpreted the aerial imagery. 
REFFERENCES 
Breiman, L., J.H. Friedman, R.A. Olshen, C.J. Stone, 
1984. Classification and Regression Trees. Wadsworth, 
Belmont. 
Cestnik, B., I. Kononenko, I. Bratko, 1987. 
ASSISTANT 86: A knowledge elicitation tool for 
sophisticated users. In: Bratko, I. and Lavrac, N., 
Progress in Machine Learning Wilmslow: Sigma Press), 
pages 31-45. 
Dunn R., A.R. Harrison, J.C. White, 1990. Positional 
accuracy and measurement error in digital databases of 
land use: an empirical study. Int. J. Geographical 
Information Systems, 1990, vol. 4, no. 4, pp. 385-398. 
European Commission - PHARE program (EC), 1997. 
Landsat TM image 191/28, acq. date 26 July 1995 for the 
CORINE Land Cover Slovenia project (EC PHARE). 
European Commission (EC), 1993. CORINE land 
cover - technical quide. European Commission, Brussels, 
1993. 
European Environment Agency (EEA), 1998. The 
Slovenian CORINE Land Cover database. Vector GIS 
database, http://etc.satellus.se. 
European Environment Agency Phare Topic Link on 
Land Cover (EEA PTL/LC), 1998. Final version of the 4th 
level CORINE land cover classes. GISAT, Prague, 1998. 
42 p. 
Gabrovec M., 1996. Solar radiation and the diverse 
relief of Slovenia. Geografski zbornik XXXVI, Ljubljana, 
1996, p. 47-68. 
Hladnik D., 1998. Large-scale monitoring of forest 
ecotopes for close-to-nature management of forests and 
forested landscape. Doctoral thesis. University of 
Ljubljana, Ljubljana, Slovenia, 1999. 286 p. 
Kobler A., M. Hocevar, 1999. Statistical analysis of the 
Slovenian CORINE Land Cover database. Unpublished. 
Slovenian Forestry Institute, Ljubljana, Slovenia. 20p. 
Kobler, A., B. Vrscaj, M. Poljak, M. Hocevar, F. Lobnik, 
1998. CORINE land cover Slovenia GIS database project 
- current status. Proc. Inti. Conf. GIS for Earth Science 
Applications. Institute for Geology, Geotechnics and 
Geophysics, Ljubljana, 1998. p. 101-110. 
McCormick N., 1997. Preprocessing and spatial 
analysis of satellite imagery using the SILVICS software 
(draft version Dec. 1997). Environmental Mapping and 
Modelling Unit, Space Applications Institute, EC's Joint 
Research Centre, Ispra, Italy. 
McCormick N., 1999. SILVICS softvare. EC's Joint 
Research Centre, Ispra, Italy, 1999. 
Michie, D., and R. Camacho, 1994. Building symbolic 
representations of intuitive real-time skills from 
performance data. In Machine Intelligence 13, (eds.K. 
Furukawa,D.Michie and S. Muggleton). Oxford University 
Press, pp. 385-418 
Ministry of Agriculture, Forestry in Food (MAFF), 1997. 
Program of forest development in Slovenia. Ministry of 
Agriculture, Forestry in Food, Ljubljana, Slovenia. 58 p. 
National Aerospace Laboratory (NLR), 1997. 
Orthorectification of the Landsat TM image 191/28, acq. 
date 26 July 1995 for the CORINE Land Cover Slovenia 
project (EC PHARE). 
Pavlin B., 1996. Primer uporabe satelitskih podatkov 
za izdelavo karte rabe tal. In: Matko D. (editor) 1999, 
Uporaba vesoljskih tehnologij. Didakta, Radovljica, 
Slovenia, 1999. 312 p. 
Perko D., M. Orozen-Adamic (editors), 1998. Slovenija 
- pokrajine in ljudje. Zalozba Mladinska knjiga, Ljubljana, 
Slovenija, 1998. 735 p. 
POP_DENSITY (Orozen) 
Quinlan, J.R. 1986. Induction of decision trees. 
Machine Learning 5(3): 239-266. 
Quinlan, J.R. 1993. Programs for Machine Learning. 
Morgan Kaufmann, San Mateo CA. 
Quinlan, J.R. 1998. C5/See5 software. 
http://www.rulequest.com/ 
Smith J. A., T.L. Lin, K.J. Ranson, 1980. The 
Lambertian assumption and Landsat data. 
Photogrammetric Engineering and remote Sensinq, 46 
(9): 1183-1189. 
Surveying and Mapping Authority of Slovenia (SMAS), 
1995. Digital elevation model, 100 m resolution. 
Surveying and Mapping Authority of Slovenia (SMAS), 
1995. Scanned topographical map 1:50.000, published 
1981. Raster GIS layer. SMAS, Ljubljana, Slovenia, 1995. 
Tretjak A., 1999. The updating of the statistical land 
cover GIS, Pomurska statistical region, from 1993 to 
1997. In: Rapid reports 4 May 1999, no. 122. Statistical 
office of the republic of Slovenia. Ljubljana, Slovenia, 
1999. 5 p. 
Wilkinson G., S. Folving, K. Fullerton, J. Megier, 1992. 
A study on the automatic revision of the European 
Community's CORINE Land Cover database using 
satellite data. Congress of the Int. Soc. for 
Photogrammetry and Remote Sensing, Washington D.C., 
August 2-14, 1992. 
Zonta I., and Surveying and Mapping Authority of 
Slovenia (SMAS), 1999. Forest mask of Slovenia - 
partially updated 1984 situation. Vector GIS database. 
SMAS, Ljubljana, Slovenian 999. 
Zumer L., 1976. Delez gozdov v slovenskem prostoru. 
Institut za gozdno in lesno gospodarstvo. Ljubljana, 
Slovenia, p. 136-139.