Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection

The land use land cover (LULC) map is extensively employed for different purposes. Machine learning (ML) algorithms applied in remote sensing (RS) data have been proven effective in image classification, object detection, and semantic segmentation. Previous studies have shown that random forest (RF)...

Full description

Saved in:
Bibliographic Details
Main Authors: Yan-Cheng Tan, Lia Duarte, Ana Cláudia Teodoro
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Land
Subjects:
Online Access:https://www.mdpi.com/2073-445X/13/11/1878
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850068676919689216
author Yan-Cheng Tan
Lia Duarte
Ana Cláudia Teodoro
author_facet Yan-Cheng Tan
Lia Duarte
Ana Cláudia Teodoro
author_sort Yan-Cheng Tan
collection DOAJ
description The land use land cover (LULC) map is extensively employed for different purposes. Machine learning (ML) algorithms applied in remote sensing (RS) data have been proven effective in image classification, object detection, and semantic segmentation. Previous studies have shown that random forest (RF) and support vector machine (SVM) consistently achieve high accuracy for land classification. Considering the important role of Portugal’s Serra da Estrela Natural Park (PNSE) in biodiversity and nature conversation at an international scale, the availability of timely data on the PNSE for emergency evaluation and periodic assessment is crucial. In this study, the application of RF and SVM classifiers, and object-based (OBIA) and pixel-based (PBIA) approaches, with Sentinel-2A imagery was evaluated using Google Earth Engine (GEE) platform for the land cover classification of a burnt area in the PNSE. This aimed to detect the land cover change and closely observe the burnt area and vegetation recovery after the 2022 wildfire. The combination of RF and OBIA achieved the highest accuracy in all evaluation metrics. At the same time, a comparison with the Normalized Difference Vegetation Index (NDVI) map and Conjunctural Land Occupation Map (COSc) of 2023 year indicated that the SVM and PBIA map resembled the maps better.
format Article
id doaj-art-2128031cb9874339803f16d0dc55cbe7
institution DOAJ
issn 2073-445X
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Land
spelling doaj-art-2128031cb9874339803f16d0dc55cbe72025-08-20T02:47:59ZengMDPI AGLand2073-445X2024-11-011311187810.3390/land13111878Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change DetectionYan-Cheng Tan0Lia Duarte1Ana Cláudia Teodoro2Department of Geography, Faculty of Arts and Humanities, University of Porto, Via Panorâmica, 4150-564 Porto, PortugalDepartment of Geosciences, Environment and Land Planning, Faculty of Sciences, University of Porto, Rua Campo Alegre, 4169-007 Porto, PortugalDepartment of Geosciences, Environment and Land Planning, Faculty of Sciences, University of Porto, Rua Campo Alegre, 4169-007 Porto, PortugalThe land use land cover (LULC) map is extensively employed for different purposes. Machine learning (ML) algorithms applied in remote sensing (RS) data have been proven effective in image classification, object detection, and semantic segmentation. Previous studies have shown that random forest (RF) and support vector machine (SVM) consistently achieve high accuracy for land classification. Considering the important role of Portugal’s Serra da Estrela Natural Park (PNSE) in biodiversity and nature conversation at an international scale, the availability of timely data on the PNSE for emergency evaluation and periodic assessment is crucial. In this study, the application of RF and SVM classifiers, and object-based (OBIA) and pixel-based (PBIA) approaches, with Sentinel-2A imagery was evaluated using Google Earth Engine (GEE) platform for the land cover classification of a burnt area in the PNSE. This aimed to detect the land cover change and closely observe the burnt area and vegetation recovery after the 2022 wildfire. The combination of RF and OBIA achieved the highest accuracy in all evaluation metrics. At the same time, a comparison with the Normalized Difference Vegetation Index (NDVI) map and Conjunctural Land Occupation Map (COSc) of 2023 year indicated that the SVM and PBIA map resembled the maps better.https://www.mdpi.com/2073-445X/13/11/1878machine learningland cover classificationobject-based image analysispixel-based image analysisrandom forestsupport vector machine
spellingShingle Yan-Cheng Tan
Lia Duarte
Ana Cláudia Teodoro
Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
Land
machine learning
land cover classification
object-based image analysis
pixel-based image analysis
random forest
support vector machine
title Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
title_full Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
title_fullStr Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
title_full_unstemmed Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
title_short Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection
title_sort comparative study of random forest and support vector machine for land cover classification and post wildfire change detection
topic machine learning
land cover classification
object-based image analysis
pixel-based image analysis
random forest
support vector machine
url https://www.mdpi.com/2073-445X/13/11/1878
work_keys_str_mv AT yanchengtan comparativestudyofrandomforestandsupportvectormachineforlandcoverclassificationandpostwildfirechangedetection
AT liaduarte comparativestudyofrandomforestandsupportvectormachineforlandcoverclassificationandpostwildfirechangedetection
AT anaclaudiateodoro comparativestudyofrandomforestandsupportvectormachineforlandcoverclassificationandpostwildfirechangedetection