Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data

Remote sensing is essential for mapping and monitoring burnt areas. Integrating Very High-Resolution (VHR) data with medium-resolution datasets like Landsat and deep learning algorithms can enhance mapping accuracy. This study employs two deep learning algorithms, UNET and Gated Recurrent Unit (GRU)...

Full description

Saved in:
Bibliographic Details
Main Authors: Sai Balakavi, Vineet Vadrevu, Kristofer Lasko
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/10/3009
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849718933223899136
author Sai Balakavi
Vineet Vadrevu
Kristofer Lasko
author_facet Sai Balakavi
Vineet Vadrevu
Kristofer Lasko
author_sort Sai Balakavi
collection DOAJ
description Remote sensing is essential for mapping and monitoring burnt areas. Integrating Very High-Resolution (VHR) data with medium-resolution datasets like Landsat and deep learning algorithms can enhance mapping accuracy. This study employs two deep learning algorithms, UNET and Gated Recurrent Unit (GRU), to classify burnt areas in the Bandipur Forest, Karnataka, India. We explore using VHR imagery with limited samples to train models on Landsat imagery for burnt area delineation. Four models were analyzed: (a) custom UNET with Landsat labels, (b) custom UNET with PlanetScope-labeled data on Landsat, (c) custom UNET-GRU with Landsat labels, and (d) custom UNET-GRU with PlanetScope-labeled data on Landsat. Custom UNET with Landsat labels achieved the best performance, excelling in precision (0.89), accuracy (0.98), and segmentation quality (Mean IOU: 0.65, Dice Coefficient: 0.78). Using PlanetScope labels resulted in slightly lower performance, but its high recall (0.87 for UNET-GRU) demonstrating its potential for identifying positive instances. In the study, we highlight the potential and limitations of integrating VHR with medium-resolution satellite data for burnt area delineation using deep learning.
format Article
id doaj-art-0dd4e64b658a4db4ae95877fde493979
institution DOAJ
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-0dd4e64b658a4db4ae95877fde4939792025-08-20T03:12:15ZengMDPI AGSensors1424-82202025-05-012510300910.3390/s25103009Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution DataSai Balakavi0Vineet Vadrevu1Kristofer Lasko2Universities Space Research Association (USRA) Science and Technology Institute, Huntsville, AL 35805, USAJames Clemens High School, Madison, AL 35756, USAGeospatial Research Laboratory, Engineer Research and Development Center, US Army Corp of Engineers, Alexandria, VA 22315, USARemote sensing is essential for mapping and monitoring burnt areas. Integrating Very High-Resolution (VHR) data with medium-resolution datasets like Landsat and deep learning algorithms can enhance mapping accuracy. This study employs two deep learning algorithms, UNET and Gated Recurrent Unit (GRU), to classify burnt areas in the Bandipur Forest, Karnataka, India. We explore using VHR imagery with limited samples to train models on Landsat imagery for burnt area delineation. Four models were analyzed: (a) custom UNET with Landsat labels, (b) custom UNET with PlanetScope-labeled data on Landsat, (c) custom UNET-GRU with Landsat labels, and (d) custom UNET-GRU with PlanetScope-labeled data on Landsat. Custom UNET with Landsat labels achieved the best performance, excelling in precision (0.89), accuracy (0.98), and segmentation quality (Mean IOU: 0.65, Dice Coefficient: 0.78). Using PlanetScope labels resulted in slightly lower performance, but its high recall (0.87 for UNET-GRU) demonstrating its potential for identifying positive instances. In the study, we highlight the potential and limitations of integrating VHR with medium-resolution satellite data for burnt area delineation using deep learning.https://www.mdpi.com/1424-8220/25/10/3009satellite remote sensingvery high resolution (VHR)PlanetScopedeep learningUNETGRU
spellingShingle Sai Balakavi
Vineet Vadrevu
Kristofer Lasko
Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
Sensors
satellite remote sensing
very high resolution (VHR)
PlanetScope
deep learning
UNET
GRU
title Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
title_full Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
title_fullStr Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
title_full_unstemmed Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
title_short Exploring Burnt Area Delineation with Cross-Resolution Mapping: A Case Study of Very High and Medium-Resolution Data
title_sort exploring burnt area delineation with cross resolution mapping a case study of very high and medium resolution data
topic satellite remote sensing
very high resolution (VHR)
PlanetScope
deep learning
UNET
GRU
url https://www.mdpi.com/1424-8220/25/10/3009
work_keys_str_mv AT saibalakavi exploringburntareadelineationwithcrossresolutionmappingacasestudyofveryhighandmediumresolutiondata
AT vineetvadrevu exploringburntareadelineationwithcrossresolutionmappingacasestudyofveryhighandmediumresolutiondata
AT kristoferlasko exploringburntareadelineationwithcrossresolutionmappingacasestudyofveryhighandmediumresolutiondata