×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Fusion of sensor data, knowledge sources and algorithms for extraction and classification of topographic objects
Author
Baltsavias, Emmanuel P.

International Archives of Photogrammetry and Remote Sensing, Vol. 32, Part 7-4-3 W6, Valladolid, Spain, 3-4 June, 1999
ANISOTROPY INFORMATION FROM MOMS-02/PRIRODA STEREO DATASETS - AN ADDITIONAL PHYSICAL
PARAMETER FOR LAND SURFACE CHARACTERISATION
Th. Schneider 1 ,1. Manakos 1 , Peter Reinartz 2 , R. Muller 2
1 Institute for Land Use Planning and Nature Conservation, University of Munich, Am Hochanger 13, D-85354 Freising, e-
mail:Tomi.Schneider@lrz.uni-muenchen.de, Ioannis.Manakos@lrz.uni-muenchen.de
2 Institute for Optoelectronics, Postfach 1116, D-82230 Wessling, e-mail: Peter.Reinartz@dlr.de, Rupert.Mueller@dlr.de
KEYWORDS: MOMS-2P, Stereo Data, Anisotropy of Backscatter, Angular Signature, BRDF, Radiometric Correction.
ABSTRACT
Multispectral and stereo band evaluations of mode D data (Blue, NIR, Pan fore and aft) from the M0MS-2P mission on MIR are
presented. Along-track MOMS-02 stereo datasets are observations of the earth surface from three distinct view directions with the
same illumination conditions. MOMS-02 mode D data allows to investigate three out of five signature types known in remote
sensing: spatial, spectral and angular signatures. The remotely sensed signal is always a combination of these three signature types.
Visual as well as computer based analysis indicates, that the result of multispectral and anisotropy based classification is determined
by different biophysical parameters. In case of vegetation, the spectral approach explores mainly the absorption of incoming radiation
by pigments and water, while the anisotropy information is due to stand structure and plant architecture. The synergistic potential of
the combined use of multispectral and anisotropy information is pointed out by the increase of the classification accuracy for the
combined multispectral and stereo bands evaluation. The results of visual interpretation led to the conclusion that a further
substantial increase can be expected using common classification routines.
For the retrieval of bio- and geo-physical parameters, the accurate radiometric correction is of decisive importance. Both, radiometric
calibration of the sensor, as well as the correction for atmospheric attenuation, are to be performed before a quantitative data
analysis. The method used for the radiometric calibration of the M0MS-2P CCD sensor lines used for multispectral and stereo data
registration is presented and the results are briefly discussed. The correction for atmospheric attenuation of the stereo data is still
under investigation. Due to differing illumination-to-sensor angle for the three view directions of a stereo dataset, each band is
corrected individually. The problem is the correct estimation of aerosol type and aerosol scattering function determination.
1. INTRODUCTION
The paper presents an approach, which first became possible
with quasi-simultaneous stereo datasets from the MOMS-02
system: the analysis of angular signatures from space. For Earth
observation, an additional physical parameter can be
investigated. Combined with the well-known multispectral
information, an increase in classification accuracy and status
assessment, as well as improvements in relating the remotely
sensed signal to bio- and geo-physical parameters of the
surfaces of interest are expected.
First investigations with a MOMS-02/D2 dataset from the
Sinaloa district in Mexico with intensive cultivated crops and
vegetables in flat terrain (Schneider et al., 1999) demonstrate
the surprisingly high potential of the anisotropy approach.
The presented paper investigates the approach in a Mid-
European, flat to hilly landscape in Bavaria, 40 km North from
Munich. Compared to the Mexico study, two key parameters are
different:
• MOMS-2P mode D consists of multispectral band 1 (blue)
instead of 3 (red) (and bands 4 (NIR), 6 and 7, which were
in both D2 and PRIRODA missions).
• the illumination-to-sensor geometry is not favourable for
angular signature extraction.
The hypotheses of the presented investigations are as follows:
• the anisotropy information of along-track stereo datasets can
be used complementarity to multispectral data analysis.
• the combined use of anisotropy and multispectral
information increase classification accuracy and status
assessment.
Problems with CCT transport from the MIR space station to the
base station in Oberpfaffenhofen, where the data were
preprocessed, have led to a delay in data delivery. Under this
circumstances the presented results are of preliminary character.
The paper focuses on features that demonstrate the potential of
the anisotropy approach by comparing results in direct
competition to the multispectral approach.
1.1. The MOMS-02 system
MOMS-02 is the first representative of a new along-track sensor
generation, designed for the combined stereo-photogrammetric
and thematic mapping of earth’s surface. The Modular
Optoelectronic Multispectral Stereo-scanner (MOMS-02) is a