Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification

Research output: Contribution to journalConference articleResearchpeer review

Authors

  • Viktoriia Hnatushenko
  • Volodymyr Hnatushenko
  • Dmytro Soldatenko

External Research Organisations

  • Ukrainian State University of Science and Technologies
  • Dnipro Polytechnic
View graph of relations

Details

Original languageEnglish
Pages (from-to)223-229
Number of pages7
JournalInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
VolumeXLVII-3-2024
Publication statusPublished - 7 Nov 2024
Event2024 Symposium on Beyond the Canopy: Technologies and Applications of Remote Sensing - Belem, Brazil
Duration: 4 Nov 20248 Nov 2024

Abstract

Today's agricultural sector is characterized by an important role of accurate mapping and monitoring of agriculture with the help of satellite imagery, which allows to optimize the use of resources, to plan crop areas and to forecast productivity. Classification of satellite images with unbalanced sample distribution is a critical problem in this regard. Traditional machine learning algorithms in particular have limitations in dealing with sample imbalance. In this paper, we proposed convolution neural networks for semantic segmentation, where sample imbalance is considered based on a particular loss function coupled with data augmentation. To illustrate our method, we use Sentinel-2 remote sensing (RS) images covering a number of regions in Ukraine, and then we create an image dataset of the region and for training and testing make data augmentation. The models with different architectural features were investigated. The results demonstrate that the proposed CNN has a higher classification accuracy than the ones discussed in the paper: the classification accuracy on the test dataset reached 96.7% with intersection-over-union values of up to 89.7%. This opens the way for further research in the direction of refining algorithms for classify satellite data with an imbalanced class structure.

Keywords

    CNN, Forest Fire, imbalanced class structure, Satellite Images

ASJC Scopus subject areas

Cite this

Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification. / Hnatushenko, Viktoriia; Hnatushenko, Volodymyr; Soldatenko, Dmytro.
In: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, Vol. XLVII-3-2024, 07.11.2024, p. 223-229.

Research output: Contribution to journalConference articleResearchpeer review

Hnatushenko, V, Hnatushenko, V & Soldatenko, D 2024, 'Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification', International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, vol. XLVII-3-2024, pp. 223-229. https://doi.org/10.5194/isprs-archives-XLVIII-3-2024-223-2024
Hnatushenko, V., Hnatushenko, V., & Soldatenko, D. (2024). Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, XLVII-3-2024, 223-229. https://doi.org/10.5194/isprs-archives-XLVIII-3-2024-223-2024
Hnatushenko V, Hnatushenko V, Soldatenko D. Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives. 2024 Nov 7;XLVII-3-2024:223-229. doi: 10.5194/isprs-archives-XLVIII-3-2024-223-2024
Hnatushenko, Viktoriia ; Hnatushenko, Volodymyr ; Soldatenko, Dmytro. / Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification. In: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives. 2024 ; Vol. XLVII-3-2024. pp. 223-229.
Download
@article{4553a64d39974a8c8af69775da79124e,
title = "Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification",
abstract = "Today's agricultural sector is characterized by an important role of accurate mapping and monitoring of agriculture with the help of satellite imagery, which allows to optimize the use of resources, to plan crop areas and to forecast productivity. Classification of satellite images with unbalanced sample distribution is a critical problem in this regard. Traditional machine learning algorithms in particular have limitations in dealing with sample imbalance. In this paper, we proposed convolution neural networks for semantic segmentation, where sample imbalance is considered based on a particular loss function coupled with data augmentation. To illustrate our method, we use Sentinel-2 remote sensing (RS) images covering a number of regions in Ukraine, and then we create an image dataset of the region and for training and testing make data augmentation. The models with different architectural features were investigated. The results demonstrate that the proposed CNN has a higher classification accuracy than the ones discussed in the paper: the classification accuracy on the test dataset reached 96.7% with intersection-over-union values of up to 89.7%. This opens the way for further research in the direction of refining algorithms for classify satellite data with an imbalanced class structure.",
keywords = "CNN, Forest Fire, imbalanced class structure, Satellite Images",
author = "Viktoriia Hnatushenko and Volodymyr Hnatushenko and Dmytro Soldatenko",
note = "Publisher Copyright: {\textcopyright} Author(s) 2024.; 2024 Symposium on Beyond the Canopy: Technologies and Applications of Remote Sensing ; Conference date: 04-11-2024 Through 08-11-2024",
year = "2024",
month = nov,
day = "7",
doi = "10.5194/isprs-archives-XLVIII-3-2024-223-2024",
language = "English",
volume = "XLVII-3-2024",
pages = "223--229",

}

Download

TY - JOUR

T1 - Neural Network-Based Analysis of Forest Fire Aftermath in Class-Imbalanced Remote Sensing Earth Image Classification

AU - Hnatushenko, Viktoriia

AU - Hnatushenko, Volodymyr

AU - Soldatenko, Dmytro

N1 - Publisher Copyright: © Author(s) 2024.

PY - 2024/11/7

Y1 - 2024/11/7

N2 - Today's agricultural sector is characterized by an important role of accurate mapping and monitoring of agriculture with the help of satellite imagery, which allows to optimize the use of resources, to plan crop areas and to forecast productivity. Classification of satellite images with unbalanced sample distribution is a critical problem in this regard. Traditional machine learning algorithms in particular have limitations in dealing with sample imbalance. In this paper, we proposed convolution neural networks for semantic segmentation, where sample imbalance is considered based on a particular loss function coupled with data augmentation. To illustrate our method, we use Sentinel-2 remote sensing (RS) images covering a number of regions in Ukraine, and then we create an image dataset of the region and for training and testing make data augmentation. The models with different architectural features were investigated. The results demonstrate that the proposed CNN has a higher classification accuracy than the ones discussed in the paper: the classification accuracy on the test dataset reached 96.7% with intersection-over-union values of up to 89.7%. This opens the way for further research in the direction of refining algorithms for classify satellite data with an imbalanced class structure.

AB - Today's agricultural sector is characterized by an important role of accurate mapping and monitoring of agriculture with the help of satellite imagery, which allows to optimize the use of resources, to plan crop areas and to forecast productivity. Classification of satellite images with unbalanced sample distribution is a critical problem in this regard. Traditional machine learning algorithms in particular have limitations in dealing with sample imbalance. In this paper, we proposed convolution neural networks for semantic segmentation, where sample imbalance is considered based on a particular loss function coupled with data augmentation. To illustrate our method, we use Sentinel-2 remote sensing (RS) images covering a number of regions in Ukraine, and then we create an image dataset of the region and for training and testing make data augmentation. The models with different architectural features were investigated. The results demonstrate that the proposed CNN has a higher classification accuracy than the ones discussed in the paper: the classification accuracy on the test dataset reached 96.7% with intersection-over-union values of up to 89.7%. This opens the way for further research in the direction of refining algorithms for classify satellite data with an imbalanced class structure.

KW - CNN

KW - Forest Fire

KW - imbalanced class structure

KW - Satellite Images

UR - http://www.scopus.com/inward/record.url?scp=85213326644&partnerID=8YFLogxK

U2 - 10.5194/isprs-archives-XLVIII-3-2024-223-2024

DO - 10.5194/isprs-archives-XLVIII-3-2024-223-2024

M3 - Conference article

AN - SCOPUS:85213326644

VL - XLVII-3-2024

SP - 223

EP - 229

JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

SN - 1682-1750

T2 - 2024 Symposium on Beyond the Canopy: Technologies and Applications of Remote Sensing

Y2 - 4 November 2024 through 8 November 2024

ER -