Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence

Annually, a considerable quantity of forest is burned on a global scale. Therefore, it is essential to obtain precise and fast information regarding the size of burned regions in order to effectively monitor the adverse consequences of wildfires. The objective of this investigation is to indicate th...

Full description

Saved in:
Bibliographic Details
Main Authors: E. O. Yilmaz, T. Kavzoglu
Format: Article
Language:English
Published: Copernicus Publications 2024-11-01
Series:ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Online Access:https://isprs-annals.copernicus.org/articles/X-5-2024/251/2024/isprs-annals-X-5-2024-251-2024.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846168771584065536
author E. O. Yilmaz
T. Kavzoglu
author_facet E. O. Yilmaz
T. Kavzoglu
author_sort E. O. Yilmaz
collection DOAJ
description Annually, a considerable quantity of forest is burned on a global scale. Therefore, it is essential to obtain precise and fast information regarding the size of burned regions in order to effectively monitor the adverse consequences of wildfires. The objective of this investigation is to indicate the effectiveness and usefulness of a deep learning (DL) architecture, such as Convolutional Neural Networks (CNNs), in the mapping of areas affected by fire, employing an eXplainable artificial intelligence (XAI) algorithm known as SHapley Additive exPlanations (SHAP) with accuracy evaluation criteria. Furthermore, this paper presents the evaluation of the Çanakkale-Kizilkeçili village wildfire. The research investigated the impacts of a variety of spectral indices, including the normalized burn index (NBR), differentiated normalized burn index (dNBR), Green-Red Vegetation Index (GRVI), simple ratio vegetation index (RVI), and normalized difference vegetation index (NDVI). At the end of the training process, the model achieved a training accuracy of approximately 0.99, with model loss values converging to approximately 0.1. The findings of the burned area identification analysis indicate that by incorporating spectral indices as supplementary information, the CNN model achieved a high level of accuracy, with an overall accuracy of 98.88% and a Kappa Coefficient of 0.98. Additionally, the SHAP technique was employed to gain insights into the output of the models. The feature importances of the spectral bands were determined through the SHAP analysis of the CNN model. Hence, the significance of the auxiliary data generated by the NBR, dNBR, and NDVI indices was identified as being the highest among the original bands and auxiliary data employed in this investigation.
format Article
id doaj-art-2aa4f04dcfb04a72952dde651eb6e9cf
institution Kabale University
issn 2194-9042
2194-9050
language English
publishDate 2024-11-01
publisher Copernicus Publications
record_format Article
series ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
spelling doaj-art-2aa4f04dcfb04a72952dde651eb6e9cf2024-11-13T11:40:48ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502024-11-01X-5-202425125710.5194/isprs-annals-X-5-2024-251-2024Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial IntelligenceE. O. Yilmaz0T. Kavzoglu1Dept. of Geomatic Engineering, Gebze Technical University, 41400, Gebze/Kocaeli, TürkiyeDept. of Geomatic Engineering, Gebze Technical University, 41400, Gebze/Kocaeli, TürkiyeAnnually, a considerable quantity of forest is burned on a global scale. Therefore, it is essential to obtain precise and fast information regarding the size of burned regions in order to effectively monitor the adverse consequences of wildfires. The objective of this investigation is to indicate the effectiveness and usefulness of a deep learning (DL) architecture, such as Convolutional Neural Networks (CNNs), in the mapping of areas affected by fire, employing an eXplainable artificial intelligence (XAI) algorithm known as SHapley Additive exPlanations (SHAP) with accuracy evaluation criteria. Furthermore, this paper presents the evaluation of the Çanakkale-Kizilkeçili village wildfire. The research investigated the impacts of a variety of spectral indices, including the normalized burn index (NBR), differentiated normalized burn index (dNBR), Green-Red Vegetation Index (GRVI), simple ratio vegetation index (RVI), and normalized difference vegetation index (NDVI). At the end of the training process, the model achieved a training accuracy of approximately 0.99, with model loss values converging to approximately 0.1. The findings of the burned area identification analysis indicate that by incorporating spectral indices as supplementary information, the CNN model achieved a high level of accuracy, with an overall accuracy of 98.88% and a Kappa Coefficient of 0.98. Additionally, the SHAP technique was employed to gain insights into the output of the models. The feature importances of the spectral bands were determined through the SHAP analysis of the CNN model. Hence, the significance of the auxiliary data generated by the NBR, dNBR, and NDVI indices was identified as being the highest among the original bands and auxiliary data employed in this investigation.https://isprs-annals.copernicus.org/articles/X-5-2024/251/2024/isprs-annals-X-5-2024-251-2024.pdf
spellingShingle E. O. Yilmaz
T. Kavzoglu
Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
title Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
title_full Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
title_fullStr Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
title_full_unstemmed Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
title_short Burned Area Detection with Sentinel-2A Data: Using Deep Learning Techniques with eXplainable Artificial Intelligence
title_sort burned area detection with sentinel 2a data using deep learning techniques with explainable artificial intelligence
url https://isprs-annals.copernicus.org/articles/X-5-2024/251/2024/isprs-annals-X-5-2024-251-2024.pdf
work_keys_str_mv AT eoyilmaz burnedareadetectionwithsentinel2adatausingdeeplearningtechniqueswithexplainableartificialintelligence
AT tkavzoglu burnedareadetectionwithsentinel2adatausingdeeplearningtechniqueswithexplainableartificialintelligence