%0 Conference Proceedings
%4 sid.inpe.br/marte2/2017/10.27.13.09.43
%2 sid.inpe.br/marte2/2017/10.27.13.09.44
%@isbn 978-85-17-00088-1
%F 59941
%T Metodologia de segmentação de variáveis geomorfométricas e emprego de lógica Fuzzy para mapeamento digital de solos na microbacia hidrográfica do Rio Santo Cristo - RS
%D 2017
%A Coelho, Fabrício Fernandes,
%A Giasson, Elvio,
%@electronicmailaddress fabricio.coelho@ufrgs.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 3415-3422
%S Anais
%1 Instituto Nacional de Pesquisas Espaciais (INPE)
%X Soil maps are sources of information for environmental impact assessment, land planning and management at different levels of scale analysis, but are very expensive to produce. Digital soil mapping techniques are alternatives to produce maps with higher scale level with lower cost. This paper proposes to test the potential of the Fuzzy logic for digital soil mapping using geographic information system (GIS) and image segmentation techniques on the geomorphometric derivations from SRTM (Shuttle Radar Topography Mission). The geomorphometric derivations from SRTM (elevation, slope, drainage distance and HAND - Height Above the Nearest Drainage) was managed in TerraView 4.2.1. computer application and the image segmentation techniques was managed in eCognition Developer 9.0 computer application. Fuzzy logic and digital soil maps was managed in ArcMap 10.4 in which the variables and the original map were related to accuracy analysis. Although with few geomorphometric derivations and few rules of Fuzzy inference, the results showed 67% in the overall accuracy and 36% in the Kappa index, close to the values showed in previous studies with others techniques. Fuzzy logic with linguistic terms presented high potential for soil class separation. Image segmentation techniques on the geomorphometric derivations presented high potential both in digital soil mapping and in traditional soil mapping.
%9 Solos e umidade do solo
%@language pt
%3 59941.pdf