Mapping of impermeable surfaces in Western Bahia using Machine Learning Algorithm

Mapping of impermeable surfaces in Western Bahia using Machine Learning Algorithm

Authors

DOI:

https://doi.org/10.21680/2447-3359.2024v10n2ID36390

Abstract

Urban impermeable surface is a relevant parameter in climate and environmental change and sustainability, and is fundamental in detecting urban environmental quality. Mapping these surfaces makes it possible to measure a city's level of urbanization, as well as generating indications of social, economic and environmental impacts. However, few studies have used high spatial resolution satellite images to analyze cities with significant urban growth in recent years, especially in western Bahia. This work aims to map the impermeable areas of the city of Barreiras–BA using CBERS 4A images and the Random Forest (RF) machine learning algorithm. The visible spectral bands obtained on 13/06/2023 were used as the basis for color compositing and image fusion to obtain 2 m pixels. The fused image was classified using the (RF) and validated using the confusion matrix, global accuracy and Kappa index. The results showed that in the city of Barreiras 41.11% correspond to impermeable surfaces. The accuracy metrics found were 0.79 for the Kappa Index and 91.7% for Global Accuracy. The results found can be used as a basis for future research into mapping land use and occupation in urban perimeters. 

Downloads

Download data is not yet available.

Author Biographies

Gustavo Gonçalves de Sousa, Universidade Federal do Oeste da Bahia

Engenheiro Civil (Centro das Ciências Exatas e das Tecnologias)

Elvis Bergue Mariz Moreira, Universidade Federal do Oeste da Bahia

Bachelor's degree (2006), Master's degree (2009), Doctorate (2014) and Post-doctorate (2016), both in Geography in the Remote Sensing line, carried out in the Postgraduate Program in Geography at the Federal University of Pernambuco PPGEO-UFPE. Taught remote sensing courses at PPGEO-UFPE at doctoral level. He is currently an Adjunct Professor at the Federal University of Western Bahia and Coordinator of the Bachelor's Degree in Geography, working in the area of ​​Geotechnology. Has experience with Remote Sensing applied to the monitoring of natural resources implemented with the Surface Energy Balance Algorithm (SEBAL), in estimating biophysical parameters, components of the radiation balance, energy and real evapotranspiration in images from the Landsat 5 TM and Landsat 8 satellites OLI/TIR.

Admilson da Penha Pacheco, Universidade Federal de Pernambuco

Degree in Physics. Master in Remote Sensing from the National Institute for Space Research (INPE) (7 Capes Concept) and PhD in Geophysics from the University of São Paulo - USP - Astronomical and Geophysical Institute - IAG (6 Capes Concept); Post-Doctorate at the Institute of Earth Sciences - University of Minho/Portugal (2019). Professor at the Federal University of Pernambuco (Center for Technology and Geosciences - Department of Cartographic Engineering and Surveying). Coordination and Participation in Research and Development Projects (CNPq, FINEPE, ANEEL, ANA and FACEPE). Areas of activity: Applied Geophysics, Environmental Sciences, Geosciences, Remote Sensing, Applied Computing and Image Processing of Natural and Artificial Materials.

Fabio Corrêa Alves, Universidade Federal do Oeste da Bahia

He holds a Bachelor's degree in Geography from the State University of Maringá - UEM (2013), a Master's degree (2015) and a PhD (2021) in Remote Sensing from the National Institute for Space Research - INPE, with a sandwich internship (2019-2020) at the University of Plymouth, England. He is an adjunct professor at the Federal University of Western Bahia - UFOB, teaching subjects and developing research in the areas of Geotechnologies. He has experience in Geoprocessing and Remote Sensing and their applications in Geosciences, working on: analysis of the physical environment, land use and land cover, geological and geomorphological investigation of fluvial landscapes, including passive continental margins and large Amazon basins, relief modeling and extraction of information from topographic metrics.

 

Henrique dos Santos Ferreira, Universidade Estadual do Piauí

Graduated, Master and Doctor in Geography from the Federal University of Pernambuco (UFPE). He works mainly in the areas of physical geography (emphasis on climatology) and geotechnologies, with an emphasis on geographics data science, remote sensing, geospatial data modeling, spatial statistics and WEB GIS development. He develops and publishes research on urban climate, climate and health, spatial dynamics of arboviruses and environmental remote sensing. He has experience in computer programming, with an emphasis on developing geospatial data analysis, time series analysis, developing interactive dashboards with the Python DASH and PLOTLY modules, modeling relational databases (SQL), geographic databases and WEB pages with HTML, CSS and JavaScript. Has experience in geospatial data analysis tools such as ArcMap, ArcGis Pro, Erdas, Qgis and the Python and JavaScript API for Google Earth Engine (GEE) and in the development of WEB GIS with ArcGis Online (ArcGis Dashboard, ArcGis APP Builder and ArcGis Experience Builder).

Downloads

Published

12-08-2024

How to Cite

SOUSA, Gustavo Gonçalves de; MOREIRA, Elvis Bergue Mariz; PACHECO, Admilson da Penha; ALVES, Fabio Corrêa; FERREIRA, Henrique dos Santos. Mapping of impermeable surfaces in Western Bahia using Machine Learning Algorithm: Mapping of impermeable surfaces in Western Bahia using Machine Learning Algorithm. Notheast Geoscience Journal, [S. l.], v. 10, n. 2, p. 141–158, 2024. DOI: 10.21680/2447-3359.2024v10n2ID36390. Disponível em: https://www.periodicos.ufrn.br/revistadoregne/article/view/36390. Acesso em: 13 sep. 2026.

Issue

Vol. 10 No. 2 (2024):

Section

Artigos

License

Copyright (c) 2024 Notheast Geoscience Journal

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Articles published by REGNE are made available under the Creative Commons Attribution 4.0 International (CC BY 4.0) license, unless expressly stated otherwise in the article itself.

This license allows anyone, free of charge and without the need for prior authorization from the journal or the authors, to read, download, copy, reproduce, distribute, share, search, data-mine, and use the articles for any lawful purpose, including commercial use, provided that proper credit is given to the authors and the original publication.

Attribution must identify, at a minimum, the authors, the title of the work, REGNE as the original publication, the year of publication, the article's DOI or permanent URL, and the indication of the CC BY 4.0 license.

The full license can be viewed at: https://creativecommons.org/licenses/by/4.0/deed.pt-br