DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network

Remote sensing image segmentation holds significant application value in precision agriculture, environmental monitoring, and other fields. However, in the task of fine-grained segmentation of weeds and crops, traditional deep learning methods often fail to balance global semantic information with l...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiangfei She, Zhankui Tang, Xin Pan, Jian Zhao, Wenyu Liu
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/13/2203
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319641882558464
author Xiangfei She
Zhankui Tang
Xin Pan
Jian Zhao
Wenyu Liu
author_facet Xiangfei She
Zhankui Tang
Xin Pan
Jian Zhao
Wenyu Liu
author_sort Xiangfei She
collection DOAJ
description Remote sensing image segmentation holds significant application value in precision agriculture, environmental monitoring, and other fields. However, in the task of fine-grained segmentation of weeds and crops, traditional deep learning methods often fail to balance global semantic information with local detail features, resulting in over-segmentation or under-segmentation issues. To address this challenge, this paper proposes a segmentation model based on a dual-branch Transformer architecture—DBFormer—to enhance the accuracy of weed detection in remote sensing images. This approach integrates the following techniques: (1) a dynamic context aggregation branch (DCA-Branch) with adaptive downsampling attention to model long-range dependencies and suppress background noise, and (2) a local detail enhancement branch (LDE-Branch) leveraging depthwise-separable convolutions with residual refinement to preserve and sharpen small weed edges. An Edge-Aware Loss module further reinforces boundary clarity. On the Tobacco Dataset, DBFormer achieves an mIoU of 86.48%, outperforming the best baseline by 3.83%; on the Sunflower Dataset, it reaches 85.49% mIoU, a 4.43% absolute gain. These results demonstrate that our dual-branch synergy effectively resolves the global–local conflict, delivering superior accuracy and stability in the context of practical agricultural applications.
format Article
id doaj-art-dbb1acd7add2441688056b25bb10eb91
institution Kabale University
issn 2072-4292
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-dbb1acd7add2441688056b25bb10eb912025-08-20T03:50:21ZengMDPI AGRemote Sensing2072-42922025-06-011713220310.3390/rs17132203DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation NetworkXiangfei She0Zhankui Tang1Xin Pan2Jian Zhao3Wenyu Liu4School of Computer Technology and Engineering, Changchun Institute of Technology, Changchun 130012, ChinaSchool of Computer Technology and Engineering, Changchun Institute of Technology, Changchun 130012, ChinaSchool of Computer Technology and Engineering, Changchun Institute of Technology, Changchun 130012, ChinaSchool of Computer Technology and Engineering, Changchun Institute of Technology, Changchun 130012, ChinaSchool of Foreign Languages and Cultures, Jilin University, Changchun 130012, ChinaRemote sensing image segmentation holds significant application value in precision agriculture, environmental monitoring, and other fields. However, in the task of fine-grained segmentation of weeds and crops, traditional deep learning methods often fail to balance global semantic information with local detail features, resulting in over-segmentation or under-segmentation issues. To address this challenge, this paper proposes a segmentation model based on a dual-branch Transformer architecture—DBFormer—to enhance the accuracy of weed detection in remote sensing images. This approach integrates the following techniques: (1) a dynamic context aggregation branch (DCA-Branch) with adaptive downsampling attention to model long-range dependencies and suppress background noise, and (2) a local detail enhancement branch (LDE-Branch) leveraging depthwise-separable convolutions with residual refinement to preserve and sharpen small weed edges. An Edge-Aware Loss module further reinforces boundary clarity. On the Tobacco Dataset, DBFormer achieves an mIoU of 86.48%, outperforming the best baseline by 3.83%; on the Sunflower Dataset, it reaches 85.49% mIoU, a 4.43% absolute gain. These results demonstrate that our dual-branch synergy effectively resolves the global–local conflict, delivering superior accuracy and stability in the context of practical agricultural applications.https://www.mdpi.com/2072-4292/17/13/2203precision agriculturefine-grained weed segmentationdual-branch Transformer architecturedynamic context aggregationlocal detail enhancementedge-aware loss
spellingShingle Xiangfei She
Zhankui Tang
Xin Pan
Jian Zhao
Wenyu Liu
DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
Remote Sensing
precision agriculture
fine-grained weed segmentation
dual-branch Transformer architecture
dynamic context aggregation
local detail enhancement
edge-aware loss
title DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
title_full DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
title_fullStr DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
title_full_unstemmed DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
title_short DBFormer: A Dual-Branch Adaptive Remote Sensing Image Resolution Fine-Grained Weed Segmentation Network
title_sort dbformer a dual branch adaptive remote sensing image resolution fine grained weed segmentation network
topic precision agriculture
fine-grained weed segmentation
dual-branch Transformer architecture
dynamic context aggregation
local detail enhancement
edge-aware loss
url https://www.mdpi.com/2072-4292/17/13/2203
work_keys_str_mv AT xiangfeishe dbformeradualbranchadaptiveremotesensingimageresolutionfinegrainedweedsegmentationnetwork
AT zhankuitang dbformeradualbranchadaptiveremotesensingimageresolutionfinegrainedweedsegmentationnetwork
AT xinpan dbformeradualbranchadaptiveremotesensingimageresolutionfinegrainedweedsegmentationnetwork
AT jianzhao dbformeradualbranchadaptiveremotesensingimageresolutionfinegrainedweedsegmentationnetwork
AT wenyuliu dbformeradualbranchadaptiveremotesensingimageresolutionfinegrainedweedsegmentationnetwork