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

Proceedings of the Symposium on Progress in Data Processing and Analysis

The solution of problems in climatology, meteorology and oceanology require global observations and
short periodical large-scale mesaurements. Low resolution scanners are therefor in operation (IFOV
1km) with a wide swath (TFOV 2000 - 6000 km) in conjunktion with IR and microwave radiometers as
well as side looking radar. (Satellites METEOSAT, METEOR, NOAA, KOSMOS-1500, ERS-1,
The rough classification of the natural ressources of the biosphere, the monitoring of environmental
phenomena as well as the analysis of lineaments to discover geomorphological structures are achie
ved by using medium resolution scanner data (IFOV 100 - 200 m; TFOV 200 - 600 km). The satelite-
borne scanners LANDSAT-MSS (US) and MSU-SK (SU) supply such data.
Local analysis of geosystem conditions demands high resolution remote sensing data (IFOV 10-50
m). High geometric resolution data of the satellite scanner systems LANDSAT-TM (US), SPOT (Fr.) and
MSU-E (SU) may be used in these operations. Many problems of environmental research additionally
require high spectral or radiometric resolution data. In such cases satellite remote sensing systems are
limited. Therefore imaging spectrometers and thermovision systems are in operation on aircraft /3/.
If the use of remote sensing data multivalent is desired, a compromise must be made.
Kondratiev wrote about an economic strategy to optimize the planning of remote sensing systems /4/.
Four groups of tasks are listed for a multivalent usable remote sensing system: Oceanology, hydrolo
gy, geology and agriculture/ forestry. The number of detailed tasks in each thematical group varies from
9 to 14. based on statistical analysis, Kondratiev was able to deduce the optimal selection of wave
length bands.
3. Geometric and Spectral Signatures and their Response to Remote
Sensing Data
Photogrammetrists and cartographers have been using airborne photos for several decades on account
of their precise reflection of geographic objects. As a result of the development of computer-aided rec
tification processes (for example RECTIMAT-C from Carl Zeiss Jena), both precision and effectivity of
land-surveying are increased. Airborne photos for this purpose are practical particularly because of their
central projection and homogeneous film emulsions. The soviet space cameras KFA-1000 and MK-4
are currently surveying high resolution photos and are expected to meet all requirements of the carto
graphers in the comming decade.
In comparison, using scanner data for the purpose of metric applications is more complicated. If a
scanner is used on aircraft, the correction of image defects will be very expensive, because of the line-
scan technique and the movement of the airplane (roll, pitch and yaw).
In thematic applications, geometric features are also very informative, but the metric precision is of se
condary importance. Feature, structure and texture may be called the geometric signatures of objects
in remote sensing data. The material and morphology of the objects are the physical causes of textu
res and structures in the images. The texture of forest areas for example derives from the line and rank
distances of the trees and the oblique illumination of the scene. Stain patterns seen in an image of cul
tivated farm areas compared with it are caused by the local and feature of sections of the field contai
ning different materials at the surface. Characteristics of textures in remote sensing images depend not
only on the surface relief but also on the angles of observation and illumination.
The interpretation of the relatively different spectral remission of objects in delimited areas can be more
detailed, if the material causes of the typical character of its remission are known. For the interaction
between spectral signature and the properties of natural objects there are given a number of qualitati
ve rules (after Soellner /5/).
(1) The more vital the vegetation is in the scene the smaller will be the spectral reflectance in the ab
sorption bands of chlorophyll and water as well as in the thermal infrared band and the heavier it will
be in the near infrared band /fig. 1.1/.
(2) If the biological activity diminishes, the amount of chlorophyll in the plants decreases first and its in-
Tluence in the absorbtion band near 0.65 ^m also decreases. The strong remission intensity in the near
infrared band remains absent until the cell-structure breaks down.
(3) The smaller the intensity of thermal emission of vegetation the greater will be the transpiration and
the larger will be the leaf area index. The vegetation is taller in.