• Home
  • Search
  • A Dynamic Effective Class Balanced Approach for Remote Sensing Imagery Semantic Segmentation of Imbalanced Data
  • Cite Icon37
  • https://doi.org/10.3390/rs15071768Copy DOI Icon

A Dynamic Effective Class Balanced Approach for Remote Sensing Imagery Semantic Segmentation of Imbalanced Data

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

The wide application and rapid development of satellite remote sensing technology have put higher requirements on remote sensing image segmentation methods. Because of its characteristics of large image size, large data volume, and complex segmentation background, not only are the traditional image segmentation methods difficult to apply effectively, but the image segmentation methods based on deep learning are faced with the problem of extremely unbalanced data between categories. In order to solve this problem, first of all, according to the existing effective sample theory, the effective sample calculation method in the context of semantic segmentation is firstly proposed in the highly unbalanced dataset. Then, a dynamic weighting method based on the effective sample concept is proposed, which can be applied to the semantic segmentation of remote sensing images. Finally, the applicability of this method to different loss functions and different network structures is verified on the self-built Landsat8-OLI remote sensing image-based tri-classified forest fire burning area dataset and the LoveDA dataset, which is for land-cover semantic segmentation. It has been concluded that this weighting algorithm can enhance the minimal-class segmentation accuracy while ensuring that the overall segmentation performance in multi-class segmentation tasks is verified in two different semantic segmentation tasks, including the land use and land cover (LULC) and the forest fire burning area segmentation In addition, this proposed method significantly improves the recall of forest fire burning area segmentation by as much as about 30%, which is of great reference value for forest fire research based on remote sensing images.

Loading PDF

Similar Papers
  • PDF
  • Research Article
  • Citations14

Impact of climate change on quantity and area of forest fires in the northern part of the Black Sea Region of Ukraine

  • Oct 29, 2017
  • Ukrainian hydrometeorological journal
  • V O Balabukh +1
  • Research Article
  • Citations15

LULC-SegNet: Enhancing Land Use and Land Cover Semantic Segmentation with Denoising Diffusion Feature Fusion

  • Dec 06, 2024
  • Remote Sensing
  • Zongwen Shi +4
  • Research Article

Forest Fire Risk Zone Mapping with Prediction of Forest Fire through Support Vector Machines (SVM) for Kothagudem, Telangana, India

  • May 30, 2025
  • Indian Forester
  • Arulrajan P +2
  • Research Article
  • Citations2

토지이용에 따른 산불발생의 영향 분석: 충청북도 사례를 중심으로

  • Aug 31, 2014
  • Journal of Korean Society of Hazard Mitigation
  • Yu Ri Kim +3
  • PDF
  • Research Article
  • Citations7

Multi-Source Fusion Enhanced Feature Segmentation in Remote Sensing Imagery

  • Oct 18, 2024
  • ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
  • Siman Wang +1
  • Research Article
  • Citations3

GIS Application in Mapping of Forest and Land Fires and Mitigation in Banjarbaru City, South Kalimantan

  • Jun 01, 2019
  • IOP Conference Series: Earth and Environmental Science
  • Fikhriyah Khairunnisa +1
  • Book Chapter
  • Citations46

Chapter 12 - Semantic scene segmentation for robotics

  • Jan 01, 2022
  • Deep Learning for Robot Perception and Cognition
  • Juana Valeria Hurtado +1
  • Research Article
  • Citations1

Transferable image synthesis for remote sensing semantic segmentation via joint reference-semantic fusion

  • Mar 01, 2026
  • Information Fusion
  • Runmin Dong +8
  • PDF
  • Research Article
  • Citations71

Clouds Classification from Sentinel-2 Imagery with Deep Residual Learning and Semantic Image Segmentation

  • Jan 10, 2019
  • Remote Sensing
  • Cheng-Chien Liu +6
  • Research Article
  • Citations30

A semantic segmentation method with category boundary for Land Use and Land Cover (LULC) mapping of Very-High Resolution (VHR) remote sensing image

  • Jan 28, 2021
  • International Journal of Remote Sensing
  • Zeyu Xu +2
  • Research Article

Assessing the impact of forest management activities on fire areas in Russia

  • Jan 01, 2024
  • BIO Web of Conferences
  • Alexey Vodolazhskiy +1
  • PDF
  • Research Article
  • Citations6

Temporal–Spatial Characteristics and Influencing Factors of Forest Fires in the Tropic of Cancer (Yunnan Section)

  • Apr 05, 2024
  • Forests
  • Haichao Xu +3
  • Research Article
  • Citations7

Self-Supervised Pretraining With Monocular Height Estimation for Semantic Segmentation

  • Jan 01, 2024
  • IEEE Transactions on Geoscience and Remote Sensing
  • Zhitong Xiong +3
  • Book Chapter
  • Citations1

Land Use and Land Cover Change of Chanthaburi Watershed Following 1999, 2006 and 2013 Floods

  • Nov 17, 2019
  • Chuti Chatewutthiprapa +2
  • Research Article
  • Citations16

A Semantic Change Detection Network Based on Boundary Detection and Task Interaction for High-Resolution Remote Sensing Images.

  • Sep 01, 2025
  • IEEE transactions on neural networks and learning systems
  • Yingjie Tang +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.