166
Chapter 18
medium wide-angle aerial photography was collected at a well-measured site
at Goonoo near Dubbo in NSW. The experiments used 120-degree aerial
photography with an anti-vignetting filter, which allowed assessment of
models rather than instrument characteristics. The Li-Strahler model with
two spectral components fitted well. At high resolution, variance dominated
the BRDF ‘mean’ signal. This scene variance is important and often
dominates a single frame requiring frame stacking to average.
To overcome the high spatial variance in the photography, Daedalus Data
were also flown for Goonoo. By flying into or at 90° to the Sun, it was
possible to sense in or across the principal plane. By taking along scan
averages of up to 1000 lines, a common Sun-sensor geometry was achieved
for each scan line and each average. The SWIR and Thermal provided good
reference BRDF channels for modeling, as they were not affected by
atmosphere.
Since then, photography and scanner data of this type, as well as video
and AVHRR data are all being collated and documented to help test the
feasibility of the proposed land cover BRDF typology, and to assess the
Australian structural typology using simple and detailed models. The
capacity and potential for these studies to help with validation and
interpretation of the global products that will come on-line in the near future
is a prime opportunity and one of particular interest to the CSIRO EOC.
7.
SEARCH FOR A STRUCTURAL TYPOLOGY
The search for consistency among photography, video data, scanner data
and even AVHRR data is being pursued in a project being undertaken by
CSIRO Information and Mathematical Sciences. The pilot for this work used
data from the Daedalus Goonoo flights with the following results.
The data for Goonoo State Forest were collected by a Daedalus or
DATM scanner (DATM stands for Daedalus Advanced Thematic Mapper
scanner) for CSIRO DWR as a test for the Li-Strahler model. Unfortunately,
there was a problem with the blue, green and red bands so the models were
originally tested using the SWIR (Short Wave InfraRed, DATM bands 9 (see
Figure 1), band 10 and Thermal band (DATM band 11).
Chapter 18
medium wide-angle aerial photography was collected at a well-measured site
at Goonoo near Dubbo in NSW. The experiments used 120-degree aerial
photography with an anti-vignetting filter, which allowed assessment of
models rather than instrument characteristics. The Li-Strahler model with
two spectral components fitted well. At high resolution, variance dominated
the BRDF ‘mean’ signal. This scene variance is important and often
dominates a single frame requiring frame stacking to average.
To overcome the high spatial variance in the photography, Daedalus Data
were also flown for Goonoo. By flying into or at 90° to the Sun, it was
possible to sense in or across the principal plane. By taking along scan
averages of up to 1000 lines, a common Sun-sensor geometry was achieved
for each scan line and each average. The SWIR and Thermal provided good
reference BRDF channels for modeling, as they were not affected by
atmosphere.
Since then, photography and scanner data of this type, as well as video
and AVHRR data are all being collated and documented to help test the
feasibility of the proposed land cover BRDF typology, and to assess the
Australian structural typology using simple and detailed models. The
capacity and potential for these studies to help with validation and
interpretation of the global products that will come on-line in the near future
is a prime opportunity and one of particular interest to the CSIRO EOC.
7.
SEARCH FOR A STRUCTURAL TYPOLOGY
The search for consistency among photography, video data, scanner data
and even AVHRR data is being pursued in a project being undertaken by
CSIRO Information and Mathematical Sciences. The pilot for this work used
data from the Daedalus Goonoo flights with the following results.
The data for Goonoo State Forest were collected by a Daedalus or
DATM scanner (DATM stands for Daedalus Advanced Thematic Mapper
scanner) for CSIRO DWR as a test for the Li-Strahler model. Unfortunately,
there was a problem with the blue, green and red bands so the models were
originally tested using the SWIR (Short Wave InfraRed, DATM bands 9 (see
Figure 1), band 10 and Thermal band (DATM band 11).