%0 Conference Proceedings
%4 sid.inpe.br/iris@1912/2005/07.20.10.46
%2 sid.inpe.br/iris@1912/2005/07.20.10.46.47
%F 7878
%T Classificação de culturas irrigadas na região de Bededouro, PE, Brasil, usando a informação polarimétrica das imagens da missão SIR-C/X-SAR
%D 1996
%A Ribeiro, Sueli Pissarra Castellari,
%A Dutra, Luciano Vieira,
%A Renno, Camilo Daleles,
%A Soares, Joao Vianei,
%B Latin-American Seminar on Radar Sensing, 1.
%C Buenos Aires, AR
%8 2-4 dec. 1996
%P 165-170
%1 CONAE/INPE/ESA/SELPER/CIDA/DI-UFPe
%X This paper investigates the use of polarimetric information extracted from SIR-C/X-SAR complex polarimetric images for crop discrimination in Bebedouro region, Pernambuco state, Brasil. In addition to widely used modulus of complex channels (amplitude)another two channels, modulus and phase from complex correlation coeficient between HH and VV complex channels were also used. The phase of correlation coeficient is a smoothed version of phase difference information between HH and VV, and was used in lieu of it. Five classes; corn, soya beans, stubble, bare soil and "caatinga" were defined. To assess the discrimination power of the extracted features confusion matrix of maximum likelyhood per point classification and Jeffreys-Matusita distance were calculated for there classes. Results showed that phase information can greatly improve the classification accuracy and the modulus of correlation coeficient carries also discrimination power.
%3 INPE 6519.pdf