Machine Learning

Use of Airborne Radar Images and Machine Learning Algorithms to Map Soil Clay, Silt, and Sand Contents in Remote Areas under the Amazon Rainforest
Digital soil mapping; soil texture; radar P-band; reference area; soil survey
Ana Carolina de S. Ferreira, Marcos Bacis Ceddia, Elias Mendes Costa, Érika F. M. Pinheiro, Mariana Melo do Nascimento, Gustavo M. Vasques
PDF Citação Projeto Pôster DOI
Use of Airborne Radar Images and Machine Learning Algorithms to Map Soil Clay, Silt, and Sand Contents in Remote Areas under the Amazon Rainforest
Training pedologist for soil mapping-Contextualizing methods and its accuracy using the project pedagogy approach
The PPA demonstrates relevance in the teaching-learning process of soil mapping. Combination of data-driven and expert knowledge methods are advised for future courses. Soil mapping using new tools involves multidisciplinary knowledge. This course can serve as a guide to meet the needs of PronaSolos
Elias Mendes Costa, Marcos Bacis Ceddia, Felipe Nascimento dos Santos, Laiz de Oliveira Silva, Igor Prata Terra de Rezende, Douglath Alves Correa Fernandes
PDF Citação Projeto Pôster DOI
Training pedologist for soil mapping-Contextualizing methods and its accuracy using the project pedagogy approach
Mapping soil properties in a poorly-accessible area
DSM tools are useful for soil sampling and mapping in areas with limited access. Non-linear models are more effective to map soil properties than linear models. GAM_scorpan model can improve soil properties predictions. GAM smooth functions can successfully map soil properties in 3-D. The high functionality of 2-D and 3-D maps benefits decision-makers.
Elias Mendes Costa, Helena Saraiva Koenow Pinheiro, Lúcia Helena Cunha dos Anjos, Robson Altiellys Tosta Marcondes, Yuri Andrei Gelsleichter
PDF Citação Projeto DOI
Mapping soil properties in a poorly-accessible area