Road extraction from remote sensing images based on attention and adaptive feature fusion

LI Yunhong; HOU Lele; CHEN Weizhong; SU Xueping; ZHANG Gang; LIU Huan; LI Yunti

doi:10.16152/j.cnki.xdxbzr.2026-03-009

您当前的位置：

首页 >

文章列表页 >

Road extraction from remote sensing images based on attention and adaptive feature fusion

更新时间：2026-07-07

- Road extraction from remote sensing images based on attention and adaptive feature fusion
- Journal of Northwest University (Natural Science Edition) Vol. 56, Issue 3, Pages: 561-571(2026)
- 作者机构：
  
  西安工程大学　电子信息学院，陕西　西安　710048
- 作者简介：
- 基金信息：
- DOI：10.16152/j.cnki.xdxbzr.2026-03-009
  CLC： TP751
- Received：28 December 2025，
  
  Revised：2026-01-10，
  
  Published：25 June 2026
- 稿件说明：
移动端阅览
LI Yunhong, HOU Lele, CHEN Weizhong, et al. Road extraction from remote sensing images based on attention and adaptive feature fusion[J]. Journal of Northwest University (Natural Science Edition), 2026, 56(3): 561-571.
DOI：

LI Yunhong, HOU Lele, CHEN Weizhong, et al. Road extraction from remote sensing images based on attention and adaptive feature fusion[J]. Journal of Northwest University (Natural Science Edition), 2026, 56(3): 561-571. DOI： 10.16152/j.cnki.xdxbzr.2026-03-009.

摘要

针对遥感图像道路提取任务中道路特征提取不充分、细节信息丢失以及类别不平衡等问题，提出了一种结合注意力与自适应特征融合（attention and adaptive feature fusion unet，AAF-Unet）的遥感图像道路提取模型。该模型以VGG16为骨干网络，首先，在Unet网络跳跃连接处加入跨空间多尺度注意力（cross-spatial efficient multi-scale attention，CEMA）模块，有效融合全局上下文信息，提升了对复杂背景和遮挡区域的特征感知能力。其次，在编码器和解码器桥接处加入自适应卷积融合（adaptive convolution mix，ACmix）模块，通过动态卷积与自注意力机制的结合，实现了局部细节特征与全局上下文信息的高效集成，显著提升了模型在边缘和细节恢复方面的性能。结合Dice和Focal设计复合损失函数，优化了模型在小目标和边缘模糊区域分割任务中的表现。最后，在不同数据集上进行实验验证，AAF-Unet模型的mIoU指标在DeepGlobe数据集上达到了82.01%，在CHN6-CUG数据集上达到了79.36%，在2个数据集上相较于Unet、Deeplabv3+、Hrnet模型均有所提升，在保证模型泛化能力的同时取得了较好的分割精度。

Abstract

To tackle the challenges of inadequate feature representation

detail degradation

and imbalanced class distribution in road extraction from remote sensing imagery

this research introduces a road extraction model from remote sensing images based on attention and adaptive feature fusion Unet (AAF-Unet). The model uses VGG16 as the backbone network. Firstly

the cross-spatial efficient multi-scale attention (CEMA) module is added at the skip connection of the Unet network

effectively integrating global contextual information and enhancing the model’s feature perception ability for complex backgrounds and occluded areas. Secondly

add the adaptive convolution mix (ACmix) module at the bridge between the encoder and the decoder. By combining dynamic convolution with self-attention mechanisms

this module enables efficient integration of local detail features and global context information

significantly improving the model’s performance in edge and detail recovery. In the loss function section a composite loss function combining Dice loss and Focal loss is proposed

optimizing the model’s performance in segmentation tasks involving small targets and edge-blurred regions. Lastly

through the experimental verification on different dataset

the mIoU index of AAF-Unet model reached 82.01% on DeepGlobe dataset and 79.36% on the CHN6-CUG dataset. Compared with Unet

Deeplabv3+and Hrnet models

the extraction results were increased respectively on two dataset. Good segmentation accuracy was achieved while ensuring the generalization ability of the model.

关键词

Keywords

references

李云红，魏小双，苏雪平，等多尺度特征增强与交互融合的遥感小目标检测 [J ] . 西北大学学报（自然科学版）， 2025 , 55 ( 2 ): 277 - 285 .

LI Y H , WEI X S , SU X P , et al . Remote sensing small target detection based on feature multi-scale enhancement and interaction fusion [J ] . Journal of Northwest University (Natural Science Edition) , 2025 , 55 ( 2 ): 277 - 285 .

CHEN R N , LI X , HU Y , et al . Road extraction from remote sensing images in wildland-urban interface areas [J ] . IEEE Geoscience and Remote Sensing Letters , 2022 , 19 : 1 - 5 .

胡宏宇，左记祥，吕颖，等 . 面向自动驾驶的遥感影像路网检测方法 [J ] . 中国公路学报， 2022 , 35 ( 11 ): 310 - 317 .

HU H Y , ZUO J X , LYU Y , et al . Road network detection of remote sensing images for autonomous driving [J ] . China Journal of Highway and Transport , 2022 , 35 ( 11 ): 310 - 317 .

ZHANG X Y , JIANG Y , WANG L Z , et al . Complex mountain road extraction in high-resolution remote sensing images via a light roadformer and a new benchmark [J ] . Remote Sensing , 2022 , 14 ( 19 ): 4729 .

ZHAO K J , LIU J , WANG Q N , et al . Road damage detection from post-disaster high-resolution remote sensing images based on TLD framework [J ] . IEEE Access , 2022 , 10 : 43552 - 43561 .

李云红，张富星，苏雪平，等 . 增强上下文特征交互的实时无人机影像分割算法 [J ] . 北京航空航天大学学报， 2026 , 52 ( 3 ): 668 - 677 .

LI Y H , ZHANG F X , SU X P , et al . Real-time UAV image segmentation algorithm with enhanced contextual feature interaction [J ] . Journal of Beijing University of Aeronautics and Astronautics , 2026 , 52 ( 3 ): 668 - 677 .

LONG J , SHELHAMER E , DARRELL T . Fully convolutional networks for semantic segmentation [C ] // 2015 IEEE Conference on Computer vision and Pattern Recognition . Boston, MA : IEEE , 2015 : 3431 - 3440 .

ZHONG Z L , LI J , CUI W H , et al . Fully convolutional networks for building and road extraction: Preliminary results [C ] // 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) . Beijing : IEEE , 2016 : 1591 - 1594 .

RONNEBERGER O , FISCHER P , BROX T . U-Net: Convolutional networks for biomedical image segmentation [M ] // Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015 . Munich : Springer International Publishing , 2015 : 234 - 241 .

佟喜峰，张婉莹 . 基于改进U-Net模型的遥感影像道路提取方法研究 [J ] . 计算机与数字工程， 2024 , 52 ( 5 ): 1495 - 1501 .

TONG X F , ZHANG W Y . Research on road extraction method from remote sensing image based on improved U-Net model [J ] . Computer & Digital Engineering , 2024 , 52 ( 5 ): 1495 - 1501 .

何哲，陶于祥，罗小波，等 . 基于改进U-Net的遥感图像道路提取 [J ] . 激光与光电子学进展， 2023 , 60 ( 16 ): 1628004 .

HE Z , TAO Y X , LUO X B , et al . Road extraction from remote sensing image based on an improved U-Net [J ] . Laser & Optoelectronics Progress , 2023 , 60 ( 16 ): 1628004 .

齐然然，帕力旦•吐尔逊，汤泊川，等 . 基于残差注意力编-解码网络的道路提取方法 [J ] . 计算机工程与科学， 2025 , 47 ( 1 ): 119 - 129 .

QI R R , PALIDAN T , TANG B C , et al . A road extraction method based on residual attention encoder-decoder network [J ] . Computer Engineering & Science , 2025 , 47 ( 1 ): 119 - 129 .

郝明，白鹤，徐婷婷 . 融合ResNeSt和多尺度特征融合的遥感影像道路提取 [J ] . 光电工程， 2025 , 52 ( 1 ): 240236 .

HAO M , BAI H , XU T T . Remote sensing image road extraction by integrating ResNeSt and multi-scale feature fusion [J ] . Opto-Electronic Engineering , 2025 , 52 ( 1 ): 240236 .

曲海成，王莹，刘腊梅，等 . 融合CNN与Transformer的遥感影像道路信息提取 [J ] . 自然资源遥感， 2025 , 37 ( 1 ): 38 - 45 .

QU H C , WANG Y , LIU L M , et al . Information extraction of roads from remote sensing images using CNN combined with Transformer [J ] . Remote Sensing for Natural Resources , 2025 , 37 ( 1 ): 38 - 45 .

唐伟，赵书河 . 基于GVF和Snake模型的高分辨率遥感图像四元数空间道路提取 [J ] . 遥感学报， 2011 , 15 ( 5 ): 1040 - 1052 .

TANG W , ZHAO S H , Road extraction in quaternion space from high spatial resolution remotely sensed images basing on GVF Snake model [J ] . Journal of Remote Sensing , 2011 , 15 ( 5 ): 1040 - 1052 .

ZHANG Z Q , BAO L Y , XIANG S , et al . B2CNet: A progressive change boundary-to-center refinement network for multi temporal remote sensing images change detection [J ] . IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2024 , 17 : 11322 - 11338 .

徐虹，杨莹洁，文武，等 . 融合双域特征均衡的遥感图像道路提取 [J ] . 电讯技术， 2024 , 64 ( 6 ): 878 - 886 .

XU H , YANG Y J , WEN W , et al . Road extraction from remote sensing images integrating dual-domain feature equalization [J ] . Telecommunication Engineering , 2024 , 64 ( 6 ): 878 - 886 .

CHEN L C , ZHU Y K , PAPANDREOU G , et al . Encoder-decoder with atrous separable convolution for semantic image segmentation [M ] // Computer Vision-ECCV 2018 . Munich : Springer International Publishing , 2018 : 833 - 851 .

ZHAO H S , SHI J P , QI X J , et al . Pyramid scene parsing network [C ] // 2017 IEEE Conference on Computer Vision and Pattern Recognition . Honolulu, HI : IEEE , 2017 : 6230 - 6239 .

WANG J D , SUN K , CHENG T H , et al . Deep high-resolution representation learning for visual recognition [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2021 , 43 ( 10 ): 3349 - 3364 .

ZHOU G D , CHEN W T , GUI Q S , et al . Split depthwise separable graph-convolution network for road extraction in complex environments from high-resolution remote-sensing images [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 5614115 .

RUAN J C , XIE M Y , GAO J S , et al . EGE-UNet: An efficient group enhanced UNet for skin lesion segmentation [C ] // Medical Image Computing and Computer Assisted Intervention-MICCAI 2023 . Cham : Springer , 2023 : 481 - 490 .

RUAN J C , LI J C , XIANG S C . VM-UNet: Vision mamba UNet for medical image segmentation [J ] . ACM Transactions on Multimedia Computing, Communications, and Applications , 2025 : 3767748 .

DENG B , LIU D , CAO Y , et al . CFRNet: Cross-attention-based fusion and refinement network for enhanced RGB-T salient object detection [J ] . Sensors , 2024 , 24 ( 22 ): 7146 .

FU Y X , LOU M , YU Y Z . SegMAN: Omni-scale context modeling with state space models and local attention for semantic segmentation [C ] // 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Nashville, IN : IEEE , 2025 : 19077 - 19087 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Prediction framework for drug-target interactions with multi-source descriptors fusion

Efficient channel attention algorithm for super-resolution reconstruction of remote sensing image

A time series signal anomaly detection method based on graph neural network

Qin Opera video denoising algorithm based on attention mechanism

Application of artificial intelligence object detection technology in disease identification of calligraphy and painting cultural relics

Related Author

DING Yanrui

CHENG Zhixing

JIANG Bo

CAI Yuhan

LI Yaowei

TONG Xiaodan

CHEN Xiaoxuan

WANG Jieting

Related Institution

Jiangsu Key Laboratory of Media Design and Software Technology, Jiangnan University

School of Science, Jiangnan University

Shaanxi Key Laboratory of Higher Education Institution of Generative Artificial Intelligence and Mixed Reality

School of Electronic Information (School of Artifical Intelligence), Northwest University

Key Laboratory of Computational Intelligence and Chinese Information Processing, Ministry of Education, Shanxi University

⁰