Fechar
Metadados

%0 Conference Proceedings
%4 sid.inpe.br/marte2/2017/10.23.19.25.18
%2 sid.inpe.br/marte2/2017/10.23.19.25.19
%@isbn 978-85-17-00088-1
%F 59610
%T Harmonização de legendas formalizadas em Land Cover Meta Language-LCML
%D 2017
%A Reis, Mariane Souza,
%A Escada, Maria Isabel Sobral,
%A Sant'Anna, Sidnei João Siqueira,
%A Dutra, Luciano Vieira,
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@electronicmailaddress reis@dpi.inpe.br
%E Gherardi, Douglas Francisco Marcolino,
%E Aragão, Luiz Eduardo Oliveira e Cruz de,
%B Simpósio Brasileiro de Sensoriamento Remoto, 18 (SBSR)
%C Santos
%8 28-31 maio 2017
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 863-870
%S Anais
%1 Instituto Nacional de Pesquisas Espaciais (INPE)
%X Many agencies have been producing land cover data with different definition, number and type of classes. In order to compare two data sets created using different legends and/or classification systems, it is necessary to carry out a harmonization process. This work aims to present and evaluate a harmonization process between two Land Cover Meta Language (LCML) formalized legends. To this purpose, two types of land cover legends obtained from maps from satellite images generated for Amazon region were analyzed and had their legends harmonized based on two approaches: 1) considering the semantic meaning of the classes and; 2) considering LCML descriptors and thresholds. The land cover maps generated with optical and SAR data were reclassified considering the harmonized legends proposed and were evaluated in terms of accuracy and agreement of the classes. For this analysis it was necessary to convert the land cover mappings to the same format and projection, and to solve problems of miss-registration. Although the harmonization process has shown to be effective, not all of the classes from TerraClass legend could be harmonized to the classes of the two maps analyzed.
%9 Processamento de imagens
%@language pt
%3 59610.pdf


Fechar