Istanbul 2004
Arable
€) A-Typel
A &-Type2
@) &Type3
A-Types
Forage
© F-Type2
Q F-Typel
@ F-Type3
-) Maize
D Trees
@ Single Trees
@ Forest
y? Roads
9 R-Type2
Q R Type!
B Houses
© Eright-roof
@ Dark-root
D Forest-Shadow
‘ognition and
rtesy of the
id Land
many)
age land varies
reason several
v samples have
ion, four levels
"ucture
es)
's only was not
d. In a second
1g class related
different levels
Shadow
on, found by
y field checks
good basis for
ng uncertainty
pot
he same image
e shown that
hierarchy have
ication results
ason for this is
alues of forage
color of some
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
types of arable land is very close to the color of roads or
building roofs. Therefore, additional information such as a more
spectral bands would be necessary in order to improve or
stabilize the result (Oesterle 2003).
4.3 Updating the LPIS
The classification result is a good source to direct an operator in
detecting changes in the LPIS data set.
The class outlines can either be exported to a GIS for interactive
work or an intersection of the classification result with the
available data might be provided to the operator. The latter one
might be a used a more convenient solution, as an operator can
superimpose the original image and the detected differences.
Figure 7: Intersection of LPIS data and classification result
The Intersection of LPIS data with the classification result is
shown in Figure 7. The black areas show differences between
the data sets to the operator. By visual checking it was found
that
l. Trees had been rooted out or planted
2. A new non-concrete road had been built
3. The field pattern changed (agricultural parcels)
The decision how to deal with these changes is in the operators
responsibility.
S. SUMMARY AND CONCLUSION
The paper investigated a variety of issues for updating a Land
Parcel Identification system. A strategy is developed and
recommendations have been drawn. These recommendations
will be shortly summarized in the following.
Orthophotos are recommended to be implemented in the LPIS.
These orthophotos can be produced, based on aerial images.
The aerial image might be recorded by airborne or spaceborne
Systems. For classification color or better multispectral images
have to be taken into account. For classification of land covers,
the season of exposure is of importance. The images have been
taken in seasons, where arable and forage land is
distinguishable. To guarantee that this holds for all images used
for updating LPIS is a certain problem with very high-resolution
satellites and maybe also for airborne remote sensing. The
problem could be minimized by taking images by systems with
519
a large swath like SPOT 5, however, the resolution of these
satellite images is not sufficient for LPIS. In orthophoto
production we have to deal with errors caused by leaning effects
and height errors of the height models used for rectification.
Using systems with a good relation of flying height and ground
coverage can reduce these errors. In this regard high resolution
satellite imagery has great benefits. Classification can help to
update the LPIS, however a detailed methodology must be
further worked out to guarantee a successful workflow.
It would be an advantage for classification, if the images are
recorded within a short period. This guarantees a nearly same
appearance (color, brightness) of land cover of the whole area.
A problem is the season of exposure is that the land cover
should be distinguishable. It is less recommended to take
images in spring, when grassland as well as arable land is
presented in green color all over.
7. REFERENCES
Definiens, 2001. eCognition User Guide Version 2.0. Edited by
Definiens Imaging GmbH; Munich, Germany
DIGITALGLOBE, — 2003. "QuickBird Imagery — Products"
Longmont Colorado
http://www .digitalelobe.com/downloads/QuickBird Imagery
Products - Product Guide.pdf
(accessed 10. June 2003)
DLR, 2002. Deutsches Zentrum für Luft- und Raumfahrt,
Institut für Weltraumsensorik und Planetenerkundung.
"High Resolution Stereo Camera - Airborne Extended HRSC-
AX", Berlin-Adlershof, Germany.
http://solarsvstem.dlr.de/FE/hrsc.shtml (accessed 10. June 2003)
EUR-Lex,, 1999. EU-Legislation "Council Regulation (EC) No
1251/1999 of 17 May 1999 establishing a support system for
producers of certain arable crops.”
http://europa.eu.int/eur-
lex/pri/en/oj/dat/1999/1 160/1_16019990626en00010014.pdf
(accessed 10. June 2003)
EUR-Lex, 1992. EU-Legislation "Council Regulation (EEC) No
3508/92 of 27 November 1992 establishing an integrated
administration and control system for certain Community aid
schemes."
http://europa.eu.int/smartapi/cei/sga doc?smartapi!celexapi!pro
d!CELEXnumdoc&lg=EN&numdoc=31992R3508& model=gui
chett
(accessed 10. June 2003)
EUR-Lex, 2000. EU - Legislation "Council Regulation (EC) No
1593/2000 of 17 July 2000 amending Regulation (EEC) No
3508/92 establishing an integrated administration and control
system for certain Community aid schemes"
http://europa.eu.int/eur-
lex/pri/en/oj/dat/2000/1_182/1_18220000721en00040007.pdf
(accessed 10. June 2003)
ImageSat International, 2003. "Satellites", Limassol, Cyprus
http://www.imagesatintl.com/aboutus/satellites/satellites.shtml£
(accessed 10. June 2003)
ITRES, 2002. "Compact airborne spectrographic Imager 2",
Calgary, Canada.
