DocumentCode :
2059972
Title :
The use of multitemporal ERS-1 SLC data for landuse classification
Author :
Kellndorfer, J. ; Schadt, R. ; Mauser, W.
Author_Institution :
Inst. of Geogr., Munchen Univ., Germany
fYear :
1993
fDate :
18-21 Aug 1993
Firstpage :
1608
Abstract :
During the commissioning phase of ERS-1 28 single-look-complex (SLC) scenes were taken of the Freiburg testsite (app. 50 km2) in the time period between Aug. 17th and Nov. 30th 1991. The scenes were converted to greyvalue images, speckle reduced using a Lee-filter and geometrically matched. A maximum-likelihood-classification of the testsite was carried out. The landuse classes were compared with a digital landuse map resulting from the 1991 ground data collection. As can be expected speckle still influences the classification results. To overcome this problem the classification results has been fit to the digital field boundary map. This enabled a fieldwise post-classification using the majority rule, which considerably improved the result of the multitemporal landuse classification
Keywords :
geophysical techniques; geophysics computing; image recognition; maximum likelihood estimation; remote sensing; remote sensing by radar; synthetic aperture radar; Lee-filter; SAR; fieldwise post-classification; geometrically matched; geophysical measurement technique; greyvalue image; land surface imaging; landuse classification; majority rule; maximum-likelihood-classification; multitemporal ERS-1 SLC; pattern recognition; radar remote sensing; single-look-complex; speckle; Geography; Image converters; Information filtering; Information filters; Layout; Management information systems; Radiometry; Remote sensing; Speckle; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium, 1993. IGARSS '93. Better Understanding of Earth Environment., International
Conference_Location :
Tokyo
Print_ISBN :
0-7803-1240-6
Type :
conf
DOI :
10.1109/IGARSS.1993.322447
Filename :
322447
Link To Document :
بازگشت