%0 Conference Proceedings
%4 sid.inpe.br/mtc-m16c/2017/12.01.21.01
%2 sid.inpe.br/mtc-m16c/2017/12.01.21.01.29
%@issn 2179-4820
%T TerraClass x MapBiomas: comparative assessment of legend and mapping agreement analysis
%D 2017
%A Neves, Alana K.,
%A Korting, Thales Sehn,
%A Fonseca, Leila Maria Garcia,
%A Queiroz, Gilberto R.,
%A Vinhas, Lubia,
%A Ferreira, Karine Reis,
%A Escada, Maria Isabel Sobral,
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%E Davis Jr., Clodoveu A. (UFMG),
%E Queiroz, Gilberto R. de (INPE),
%B Simpósio Brasileiro de Geoinformática, 18 (GEOINFO)
%C Salvador
%8 04-06 dez. 2017
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 295-300
%S Anais
%X In this work, we evaluated the agreement between land cover maps generated by TerraClass and MapBiomas projects for Pará state and, more specifically: (1) describe the legends based on an international classification system, (2) analyze the differences among classes and (3) test PostGIS Raster from PostgreSQL database to work with classification products. The classifications were compared pixel by pixel and the evaluation was performed based on confusion matrices. The agreement between them was 84.40%. The different methodologies adopted by the two projects generate significant disagreements in class identification, so using both maps together as complementary is not recommended for land use and cover change analyzes.
%@language pt
%3 37neves_escada.pdf