andslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaoca, southeastern Brazi

Author(s)

Dias, Helen Cristina ^{[1, 2]} ; Sandre, Lucas Henrique ^[3] ; Satizabal Alarcon, Diego Alejandro ^[4] ; Grohmann, Carlos Henrique ^{[1, 2]} ; Quintanilha, Jose Alberto ^[1]

Affiliation

^[1] Univ Sao Paulo, Inst Energy & Environm, Sao Paulo, SP - Brazil

^[2] Univ Sao Paulo, Inst Energy & Environm, Spatial Anal & Modelling Lab, Sao Paulo, SP - Brazil

^[3] Univ Estadual Campinas, Inst Geosci, Campinas, SP - Brazil

^[4] Univ Sao Paulo, Inst Geosci, Sao Paulo, SP - Brazil

• Document type: Journal article
• Source: BRAZILIAN JOURNAL OF GEOLOGY; v. 51, n. 4 2021.
• Web of Science Citations: 0

Abstract

Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaoca (Sao Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m(2). The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution. (AU)

• FAPESP's process: 19/26568-0 - High-resolution remote sensing, deep learning and geomorphometry in analyses of mass movements and geological risk
• Grantee: Carlos Henrique Grohmann de Carvalho
• Support Opportunities: Regular Research Grants
• FAPESP's process: 19/17261-8 - Analysis of manual and semi-automatic shallow landslides inventories and its suitability in predictive models
• Grantee: Helen Cristina Dias
• Support Opportunities: Scholarships in Brazil - Doctorate