f-the-art
zes
ficult to
| Section
) all the
rescale,
ts It can
fication,
ness of
ocessing
wavelets
rization,
f-the-art
results:
4 bands
tivariate
Factor
med by
images,
nd Line
rmed by
ies like
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B3, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
Object Based Image Analysis or joint GIS-Image analysis are
also available.
Figure 4. Example of Orfeo ToolBox algorithm results: Pan-
sharpening and Line segment Detection on a Quickbird image
of Toulouse, France.
3.3 Support of Pleiades imagery
Supporting the new Pleiades imagery in the Open source world
was a great challenge for the OTB team, since the JPEG2000
format in which images are delivered is not yet commonplace,
and since they are uncommonly huge: a standard product is 20
by 20 kilometers, sampled at 50 centimeters, with four spectral
bands. To tackle this issue, CNES and the OTB team got
involved in OpenJPEG (OpenJPEG, www.openjpeg.org ), one
of the most advanced JPEG2000 Open source alternative. In
cooperation with the OpenJPEG community, the software has
been improved a lot and is now able to handle Pleiades imagery
with reasonable performances. Additional contributions are to
be made to OSSIM in order to share the support for Pleiades
sensor modelling.
The result of all this work is that one can smoothly navigate in a
Pleiades image through Monteverdi or the standalone OTB
viewer, including access to the intermediate resolutions thanks
to JPEG2000 features. The navigation experience on a decent
computer will be rather nice, as shown in Figure 2. In addition
to navigation, one can easily uncompressed parts or totality of a
Pleiades image in reasonable time. Using Pleiades image as
input to OTB or applications plugins processing chain, though
on-the-flow decompression might become time-consuming for
- some algorithms, in which case prior decompression to disk is
advised.
4. CONCLUSION
The actual capabilities and performances of the Pleiades system
will be intensively tested during the thematic commissioning
year. The very promising results already obtained in the
calibration phase let think that images and system performances
will be of excellent quality.
Meanwhile, three other key challenges have to be tackled by
ORFEO team, both thematic and methodological. First, the
large number of studies should demonstrate and concretely
assess the benefit of sub-metric optical data for a large range of
public sector users. Second, the Orfeo ToolBox Open source
library, OTB applications and Monteverdi, should bring a real
benefit to a large number of users (from C++ developers to end-
$53
users), to use, manipulate and process these huge and incredibly
rich images. Last, the ORFEO program should significantly and
concretely demonstrate that Pleiades imagery is of great
technical benefit for both public and commercial sectors. The
final objective to reach is to show that such imagery can be
efficiently integrated into operational processes, in order that
decision makers include this new type of data into their usual
tools.
5. REFERENCES
Christophe, E. and Inglada, J., 2009: Open source remote
sensing: Increasing the usability of cutting-edge algorithms,
IEEE Geoscience and Remote Sensing Newsletter
Inglada, J. and Christophe, E., 2009 : The Orfeo Toolbox
remote sensing image processing software, Geoscience and
Remote | Sensing | Symposium | (IGARSS), 2009 IEEE
International,, Vol 4, pp IV-733
Nielsen, A.A. and Hecheltjen, A. and Thonfeld, F. and Canty,
M.J., 2010: Automatic change detection in RapidEye data using
the combined MAD and kernel MAF methods, Geoscience and
Remote | Sensing | Symposium | (IGARSS), 2010 IEEE
International, pp 3078-3081
OTB Development Team, 2011a: The ORFEO Tool Box
Software Guide.
http://www.orfeo-toolbox.org/otb/documentation.html
OTB Development Team, 2011b: The Orfeo ToolBox
Cookbook, a guide for non-developers
http://Awww.orfeo-toolbox.org/otb/documentation.html
von Gioi, R.G. and Jakubowicz, J. and Morel, J.M. and Randall,
G., 2010: Lsd: A fast line segment detector with a false
detection control, Pattern Analysis and Machine Intelligence,
IEEE Transactions, 32(4), pp 722-732
