Remote Sensing and Urban Statistics: ‚Delimitation of the Urban Agglomeration of Budapest - Test of 
Delineating Enumeration Districts in the Zuglö District (Budapest) 
Birö, M., Petrik, O., Winkler, P. 
Institute of Geodesy, Cartography and Remote Sensing (FÖMI) 
Faluvégi, A. 
Hungarian Central Statistical Office (HCSO) 
KEY WORDS: Remote Sensing, Urban Agglomeration, High Resolution Satellite Images, Digital Cadastral Database 
ABSTARCT: 
An operational method was developed/adapted to compiling land use maps at the scale 1:50.000 for the area of Budapest, at reference 
date 1996, according to framework, developed for remote sensing applications in statistics respecting statistical classification principles 
and statistical nomenclature of land use categories. The limits of the Budapest-Agglomeration were identified with the help of Earth 
Observation data according to standard European rules for these data. The possibilities of using Earth Observations data were tested in 
the process of delineation of Enumeration districts in the district of Zugló (Budapest). The method of accurate and successful delineation 
of the land use map of the Budapest agglomeration was established by means of commercially available high resolution satellite images, 
in compliance with CLUSTERS II level of Eurostat. The classification of Zugló - (XIVth district of Budapest) — was carried out 
according to the full CLUSTERS IV level. For the area of Zugló, revision of enumeration districts of the 2001 census was executed on 
the basis of satellite images. Categories for which delimitation based on satellite images seems possible, were defined. In view of 
standard interpretation, category concepts were defined in common and, furthermore, agreement was found on the principle of 
interactive, direct consultations concerning any problem occurring in the course of the forthcoming tasks. 
1. INTRODUCTION 
The project consist of the following phases (Bíró et al, 1998) 
Phase I.: elaboration of land use maps on a 1:50.000 scale and 
according to CLUSTERS nomenclature level Il 
(EUROSTAT, 1996) for the area of Budapest 
Agglomeration for the reference year 1996, quality 
control 
Phase ll: delimitation of urban morphological zone and 
identification of the urban agglomeration using results 
of Phase 1., application of method 
Phase 111. elaboration of land use maps on a 1:10.000 scale and 
according to CLUSTERS nomenclature level for the 
area of the Zugló district and the reference year 1996, 
test of delineation of town-planning district ‚and 
enumeration districts, quality control 
Phase IV.: presentation of results and deliverables 
2. MATERIALS AND METHODS 
2.1 Minimum observation unit 
The smallest interpretable observation unit is 120m x 120m, 
which seems applicable for the land use classification in the 
case of the Budapest agglomeration with a scale of 1:50.000. 
60mx60m was used in the case of the interpretation of Zugló 
district's land-use data, too 
2.2 Technology of land cover classification 
In the selection of the technology to be used, the FÓMI - in 
accordance with the HCSO - made efforts to allow the most 
efficient, accurate and reliable processing of satellite images 
possible and, at the same time, to make it possible in the future to 
extend and carry out this new task to the total area of the country. 
Bearing this in mind, we decided to use the technology of visual 
photo-interpretation of satellite images, according to the 
Statistical Framework for RS application in statistics, including 
the CLUSTERS nomenclature. Its advantage is that in the Remote 
Sensing Centre of FÓMI there are well-trained photo interpreters, 
operators, photographic laboratory technicians, having 
participated in the high standard activity of the team in setting up 
the CORINE database during the past three years, under quality 
control. . 
Ensuing to the geometric transformation of satellite images, 
several specimen copies were produced at the scale 1:50.000, 
ending with the series of satellite image maps of rectified IRS-1C 
images with resolution of 24 m. For film writing the equipment 
COLORMATION was used. The photo interpretation was carried 
out on the satellite image maps in the scale 1:50 000. The full 
coverage of the whole agglomeration of Budapest is settled by 6 
space-photo maps at this scale. 
The result of the interpretation was registered on transparencies 
superposed on space-photo maps and on a PC monitor, 
according to CLUSTERS Il. In the course of the interpretation 
topographic maps and, when necessary, aerial photos were 
used. As a next step, the reliability of the interpretation was 
checked by field inspection, and doubtful classes were clarified. 
This was followed by inking of the result of the interpretation 
on the transparencies, scanning and vectorization. 
The vector map containing CLUSTERS classes has been 
projected by means of ArcInfo and ERDAS software onto the 
space-photo and category limits were interactively adjusted. As 
a final product, a land cover map complying with CLUSTERS 
classes was supplied. The total database was transmitted both in 
digitized and analogous form to the Department of 
Photogrammetry of the Technical University for quality 
control. 
Parallel to this we carried out the classification for the Zugló 
district according to CLUSTERS IV on the base of the merged 
multispectral Indian IRS-1C and Russian KVR-1000 (2 m 
78 International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 
resol 
the § 
enun 
FÖM 
perio 
to m 
infor 
same 
a ne 
appli 
by ac 
2.3 ( 
Duri 
the 
requi 
imag 
of a 
satis 
mete 
Fren 
inter 
vege 
midc 
The 
satel 
tom 
and 
spec 
spati 
x 30 
IC i 
banc 
For 
com 
m ai 
info: 
reso. 
imag