|
METHODOLOGY FOR ESTIMATING THE PROBABILITIES OF CORRECT AND ERRONEOUS DECISIONS IN THE CLASSIFICATION OF THE SPATIALLY DISTRIBUTED TARGET OVER RADAR IMAGES
|
|
|
N. Vinogradova;L. Dorosinsky
|
|
|
||
|
|
|
|
1314-2704
|
|
|
||
|
English
|
|
|
20
|
|
|
2.2
|
|
|
|
|
|
||
|
The problem of classifying spatially distributed targets using their radar images during
the process of remote sensing has been and remains one of the main tasks when observing the Earth with space-based observation means based on the use of a synthetic aperture radar (SAR). The applying of radar monitoring means is especially relevant for the northern latitudes, for example, the territory of Russia, where, on the one hand, a rather high percentage of cloud cover, on the other hand, the duration of daylight hours is quite short for half a year. The problem of obtaining radar images has been fully studied, the algorithms for their formation have also been known for a long time, nevertheless, the evaluation of the effectiveness of such means requires careful and detailed consideration, which this work is devoted to. The exact formulas are given in the paper, which allows, based on analytical methods, to calculate the probability of correct recognition of spatially distributed objects that differ in size and distribution by scattering cross-section value on the target surface. In addition, a detailed analysis of approximate formula expressions is given, as well as the boundaries of their applicability. Examples of the application of the proposed algorithm for specific situations presented on a variety of radar portraits of objects of the earth's surface are given. Specific recommendations that take into account the various possible situations that a researcher may encounter are given. |
|
|
conference
|
|
|
||
|
||
|
20th International Multidisciplinary Scientific GeoConference SGEM 2020
|
|
|
20th International Multidisciplinary Scientific GeoConference SGEM 2020, 18 - 24 August, 2020
|
|
|
Proceedings Paper
|
|
|
STEF92 Technology
|
|
|
International Multidisciplinary Scientific GeoConference-SGEM
|
|
|
SWS Scholarly Society; Acad Sci Czech Republ; Latvian Acad Sci; Polish Acad Sci; Russian Acad Sci; Serbian Acad Sci & Arts; Natl Acad Sci Ukraine; Natl Acad Sci Armenia; Sci Council Japan; European Acad Sci, Arts & Letters; Acad Fine Arts Zagreb Croatia; C
|
|
|
241-250
|
|
|
18 - 24 August, 2020
|
|
|
website
|
|
|
cdrom
|
|
|
7068
|
|
|
radar image; spatially distributed targets; signal class recognition
|
|