The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008
4. DISCUSSION
The complexity of the pre-processing of dual-view FIGOS data
prior to BRF retrieval is mainly due to the characteristics of the
different instruments involved. Ideally, the irradiance should be
measured at the same temporal and spectral resolution and
wavelength range as the reflected radiance. Thus, a device that
could capture direct and angularly resolved diffuse irradiance at
a sampling rate equal to the target reflectance acquisition time
would be highly desirable. This would provide a complete
angular characterisation of the irradiance distribution for each
spectrodirectional target measurement.
The use of several instruments requires intercalibrations. The
storage of according intercalibration factors in the spectral
database enables the automation of intercalibration. Similarly,
the storage of Spectralon characteristics, i.e. factors describing
deviations from the ideal Lambertian . reflector, allows
correcting measurements for these imperfections (Hüni et al.,
2008). While Spectralon factors can be considered part of the
metadata space of spectral data, the same does not strictly apply
to intercalibration factors. The latter are rather metadata of the
instruments. However, their storage within a spectral database
is important as they can tie spectra to some common reference
instrument. Such capability is highly desirable when dealing
with campaigns involving many different instruments, e.g.
round robin experiments as planned in the Hyper-I-Net project
(Nieke et al., 2007).
The concept of the Space Processing Chain is using the
SPECCHIO database as data source to build the chain input
spaces. All spaces are then held in memory, leaving the original
information in the database untouched. Chain outputs are not
stored in the SPECCHIO database but can be exported to files
or stored in specialised reference databases (Hueni et al., 2008).
Due to the flexible and fast processing capabilities of the
SPECCHIO Space Chain, reprocessing of original data is far
easier than managing products by keeping track of all involved
module parameters. It is however foreseen to implement the
storage of chain configurations in the database or in
configuration files. This will enable users to store and reload
typical chain settings.
5. CONCLUSIONS
The automated pre-processing of dual-view FIGOS data is an
important step towards an operational BRF retrieval. The
utilisation of a database combined with a flexible, configurable
processing chain allows dealing with the complex processing
needs arising from instrument intercalibrations and differing
spectral and temporal resolutions of the field data sets.
The generic, modular approach to processing of spectral data
will enable the application of processing components to datasets
acquired with different sensors, thus making the system useful
for other research groups.
The inclusion of instrument intercalibration data in the database
model is enabling the SPECCHIO system to be used in round
robin experiments and is an important step towards comparable
datasets in multi-instrument campaigns and, ultimately, better
data quality.
REFERENCES
Analytical Spectral Devices Inc. 2007. Technical Guide.
Boulder, CO: Analytical Spectral Devices Inc.
Beisl, U., 2001. Correction of Bidirectional Effects in Imaging
Spectrometer Data, Remote Sensing Series No. 37. Zurich,
Remote Sensing Laboratories, pp 188
Hueni, A., Biesemans, J., Meuleman, K., Dell'Endice, F.,
Schlapfer, D., Adriaensen, S., Kempenaers, S., Odermatt, D.,
Kneubuhler, M. & Nieke, J., 2008. Structure, Components and
Interfaces of the APEX Processing and Archiving Facility.
IEEE Transactions on Geoscience and Remote Sensing,
Submitted.
Hueni, A. & Tuohy, M., 2006. Spectroradiometer Data
Structuring, Pre-Processing and Analysis - An IT Based
Approach. Journal of Spatial Science, 51(2), 93-102.
Huni, A. & Kneubuhler, M., 2007. SPECCHIO: a system for
storing and sharing spectroradiometer data. SPIE
NewsroomfDzczvNoex 2007).
Huni, A., Nieke, J., Schopfer, J., Kneubuhler, M. & Itten, K.
2007a. 2nd Generation of RSL's Spectrum Database
"SPECCHIO". 10th Inti. Symposium on Physical
Measurements and Spectral Signatures in Remote Sensing,
Davos (CH), Eds. M. E. Schaepman, S. Liang, N. E. Groot & M.
Kneubuhler. Vol. XXXVI, Part 7/C50, 505-510.
Hiini, A., Nieke, J., Schopfer, J., Kneubuhler, M. & Itten, K.
2007b, 23-25 April 2007. Metadata of Spectral Data
Collections. 5th EARSeL Workshop on Imaging Spectroscopy,
Bruges, Belgium.
Hiini, A., Nieke, J., Schopfer, J., Kneubuhler, M. & Itten, K.,
2008. The spectral database SPECCHIO for improved long
term usability and data sharing. Computers & Geosciences, .
Accepted for publication.
Labsphere. Spectralon Care and Handling Guidelines. North
Sutton, NH 03260: Labsphere.
Landgrebe, D., 1997. On Information Extraction Principles for
Hyperspectral Data. West Lafayette, IN, Purdue University, pp
34
Martonchik, J. V., 1994. Retrieval of Surface Directional
Reflectance Properties Using Ground Level Multiangle
Measurements. Remote Sens. Environ., 50, 303-316.
Nicodemus, F. E., Richmond, J. C., Hsia, J. J., Ginsberg, I. W.
& Limperis, T. 1977. Geometrical Considerations and
Nomenclature for Reflectance. Washington D.C., USA:
Institute for Basic Standards, National Bureau of Standards.
Nieke, J., Dell'Endice, F., Hiini, A., Kneubuhler, M., Schlapfer,
D., Kotz, B., Schopfer, J., Itten, K. & Plaza, A. 2007.
Calibration and Validation Activities in the Scope of HYPER-I-
NET: The RSL Approach. IEEE International Geoscience and
Remote Sensing Symposium (IGARSS'07), Barcelona, Spain.
Nieke, J., Itten, K., Debryun, W. & and the APEX Team. 2005.
The Airborne Imaging Spectrometer APEX: From Concept to