Detecting atypical agricultural patterns in rural credit applications based on satellite image time series
by SILVA, Baggio Luiz de Castro e1; FERREIRA, Karine Reis1; QUEIROZ, Gilberto Ribeiro de1; ADAMI, Marcos1; KÖRTING, Thales Sehn1.
1National Institute for Space Research (INPE)
XXI Brazilian Remote Sensing Symposium: https://proceedings.science/sbsr-2025/trabalhos/detectando-padroes-atipicos-de-agricultura-em-aplicacoes-de-credito-rural-basead
Publisher: Proceedings XXI Brazilian Remote Sensing Symposium | Published on 13 April 2025
Abstract
This paper presents a methodology for detecting atypical patterns in agricultural rural credit applications, based on time series from satellite images. The methodology employs clustering methods, specifically Self-Organizing Maps (SOM) and Hierarchical clustering, using Euclidean and Dynamic Time Warping (DTW) distances. The proposed methodology was applied to soybean fields obtained from the Rural Credit Operations System and Proagro (Sicor), obtaining promising results presented in this work.
Keywords: remote sensing, image time series, unsupervised learning, atypical patterns, agriculture
Share and Cite
SILVA, Baggio Luiz de Castro e et al. DETECTANDO PADRÕES ATÍPICOS DE AGRICULTURA EM APLICAÇÕES DE CRÉDITO RURAL BASEADO EM SÉRIES TEMPORAIS DE IMAGENS DE SATÉLITES. In: Proceedings XXI Brazilian Remote Sensing Symposium, 2025, Salvador. Galoá, 2025. Available in: <https://proceedings.science/sbsr-2025/trabalhos/detectando-padroes-atipicos-de-agricultura-em-aplicacoes-de-credito-rural-basead?lang=pt-br> Access: 2025, May 5th.