A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance

Fire poses a serious threat to the global economy, environment, and social stability, highlighting the need for rapid and accurate fire detection. Remote sensing combined with deep learning has outperformed traditional fire assessment methods. However, in early fire stages, small flame areas, class...

Full description

Saved in:
Bibliographic Details
Main Authors: Rui Zou, Zhihui Xin, Guisheng Liao, Penghui Huang, Rui Wang, Yuhu Qiao
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/13/2175
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849428693270659072
author Rui Zou
Zhihui Xin
Guisheng Liao
Penghui Huang
Rui Wang
Yuhu Qiao
author_facet Rui Zou
Zhihui Xin
Guisheng Liao
Penghui Huang
Rui Wang
Yuhu Qiao
author_sort Rui Zou
collection DOAJ
description Fire poses a serious threat to the global economy, environment, and social stability, highlighting the need for rapid and accurate fire detection. Remote sensing combined with deep learning has outperformed traditional fire assessment methods. However, in early fire stages, small flame areas, class imbalance, and weak feature extraction hinder detection accuracy. This study proposes an end-to-end segmentation model called Flame Detail Enhancement U-Net (FDE U-Net), using Landsat-8 multispectral remote sensing data. The model incorporates the self-Attention and Convolutional mixture (ACmix) module and the Convolutional Block Attention Module (CBAM) into the encoder of the Residual U-Net. ACmix integrates self-attention and convolution to capture global semantic features while maintaining computational efficiency, improving both contextual awareness and local detail. CBAM enhances flame recognition by weighting important channel features and focusing spatially on small flame areas, helping address the class imbalance problem. Additionally, Haar wavelet downsampling is applied to retain image detail and improve the detection of small-scale flame regions. Experimental results show that the FDE U-Net model exhibits robust performance in fire detection, accurately extracting flame regions even when their proportion is low and the background is complex. The F1 score reaches 95.97%, significantly improving the class imbalance problem.
format Article
id doaj-art-4c0b3f2ea7494d5eb697d9e791dec793
institution Kabale University
issn 2072-4292
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-4c0b3f2ea7494d5eb697d9e791dec7932025-08-20T03:28:37ZengMDPI AGRemote Sensing2072-42922025-06-011713217510.3390/rs17132175A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category ImbalanceRui Zou0Zhihui Xin1Guisheng Liao2Penghui Huang3Rui Wang4Yuhu Qiao5School of Physics and Electronic Information Technology, Yunnan Normal University, Kunming 650500, ChinaSchool of Physics and Electronic Information Technology, Yunnan Normal University, Kunming 650500, ChinaNational Laboratory of Radar Signal Processing, Xidian University, Xi’an 710071, ChinaSchool of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaSchool of Physics and Electronic Information Technology, Yunnan Normal University, Kunming 650500, ChinaSchool of Physics and Electronic Information Technology, Yunnan Normal University, Kunming 650500, ChinaFire poses a serious threat to the global economy, environment, and social stability, highlighting the need for rapid and accurate fire detection. Remote sensing combined with deep learning has outperformed traditional fire assessment methods. However, in early fire stages, small flame areas, class imbalance, and weak feature extraction hinder detection accuracy. This study proposes an end-to-end segmentation model called Flame Detail Enhancement U-Net (FDE U-Net), using Landsat-8 multispectral remote sensing data. The model incorporates the self-Attention and Convolutional mixture (ACmix) module and the Convolutional Block Attention Module (CBAM) into the encoder of the Residual U-Net. ACmix integrates self-attention and convolution to capture global semantic features while maintaining computational efficiency, improving both contextual awareness and local detail. CBAM enhances flame recognition by weighting important channel features and focusing spatially on small flame areas, helping address the class imbalance problem. Additionally, Haar wavelet downsampling is applied to retain image detail and improve the detection of small-scale flame regions. Experimental results show that the FDE U-Net model exhibits robust performance in fire detection, accurately extracting flame regions even when their proportion is low and the background is complex. The F1 score reaches 95.97%, significantly improving the class imbalance problem.https://www.mdpi.com/2072-4292/17/13/2175FDE U-Netfeature fusionfire segmentationmultispectral remote sensing images
spellingShingle Rui Zou
Zhihui Xin
Guisheng Liao
Penghui Huang
Rui Wang
Yuhu Qiao
A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
Remote Sensing
FDE U-Net
feature fusion
fire segmentation
multispectral remote sensing images
title A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
title_full A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
title_fullStr A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
title_full_unstemmed A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
title_short A Fire Segmentation Method with Flame Detail Enhancement U-Net in Multispectral Remote Sensing Images Under Category Imbalance
title_sort fire segmentation method with flame detail enhancement u net in multispectral remote sensing images under category imbalance
topic FDE U-Net
feature fusion
fire segmentation
multispectral remote sensing images
url https://www.mdpi.com/2072-4292/17/13/2175
work_keys_str_mv AT ruizou afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT zhihuixin afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT guishengliao afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT penghuihuang afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT ruiwang afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT yuhuqiao afiresegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT ruizou firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT zhihuixin firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT guishengliao firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT penghuihuang firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT ruiwang firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance
AT yuhuqiao firesegmentationmethodwithflamedetailenhancementunetinmultispectralremotesensingimagesundercategoryimbalance