地球信息科学学报 ›› 2022, Vol. 24 ›› Issue (5): 949-961.doi: 10.12082/dqxxkx.2022.210597

• 遥感科学与应用技术 • 上一篇    下一篇

面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型

陈振1,2,3(), 陈芸芝1,2,3,*(), 吴婷1,2,3, 李佳优1,2,3   

  1. 1.福州大学 数字中国研究院(福建),福州 350108
    2.卫星空间信息技术综合应用国家地方联合工程研究中心,福州 350108
    3.福州大学 空间数据挖掘与信息共享教育部重点实验室,福州 350108
  • 收稿日期:2021-09-30 修回日期:2021-12-03 出版日期:2022-05-25 发布日期:2022-07-25
  • 通讯作者: * 陈芸芝(1982— ),女,福建福州人,博士,副研究员,主要研究方向为环境遥感和林业遥感等。 E-mail: chenyunzhi@fzu.edu.cn
  • 作者简介:陈 振(1996— ),男,福建南平人,硕士生,主要研究方向为遥感影像信息提取。E-mail: n195527003@fzu.edu.cn
  • 基金资助:
    国家自然科学基金项目(42071446);联通(福建)产业互联网有限公司委托项目(JC21-3502-2020-000559)

A Lightweight Dual Attention and Feature Compensated Residual Network Model for Road Extraction from High-Resolution Remote Sensing Images

CHEN Zhen1,2,3(), CHEN Yunzhi1,2,3,*(), WU Ting1,2,3, LI Jiayou1,2,3   

  1. 1. The Academy of Digital China(Fujian), Fuzhou University, Fuzhou 350108, China
    2. National and Local Joint Engineering Research Center for the Comprehensive Application of Satellite Space Information Technology, Fuzhou 350108, China
    3. Key Laboratory of Spatial Data Mining and Information Sharing, Ministry of Education, Fuzhou University, Fuzhou 350108, China.
  • Received:2021-09-30 Revised:2021-12-03 Online:2022-05-25 Published:2022-07-25
  • Supported by:
    National Natural Science Foundation of China(42071446);Commissioning Project of Unicom (Fujian) Industrial Internet Co., Ltd.,(JC21-3502-2020-000559)

摘要:

针对高分辨率遥感影像背景复杂,道路提取容易受阴影、建筑物和铁路等背景信息干扰的问题,提出一种带有轻量级双注意力和特征补偿机制的DAFCResUnet模型。该模型在ResUnet的基础上,通过增加轻量级的双注意力和特征补偿模块实现模型在性能和时空复杂度上的平衡。其中,双注意力模块可以增强模型的特征提取能力,特征补偿模块可以融合网络中来自深浅层的道路特征。在DeepGlobe和GF-2道路数据集上的实验结果表明,DAFCResUnet模型的IoU和F1-score可以达到0.6713、0.8033和0.7402、0.8507,模型的整体精度优于U-Net、ResUnet和VNet模型。与U-Net和ResUnet模型相比,DAFCResUnet模型仅增加了少量的计算量和参数量,但IoU和F1-score均有较大幅度的提高;与VNet模型相比,DAFCResUnet模型在计算量和参数量远低于VNet的情况下取得了更高的精度,模型在精度和时空复杂度两方面均有优势。相比其他对比模型,DAFCResUnet模型具有更强的特征提取和抗干扰能力,能更好解决道路上的干扰物、与道路特征相似地物、树荫或阴影遮挡等造成的道路空洞、误提和漏提现象。

关键词: 深度学习, 道路提取, 高分辨率遥感影像, 残差网络, U-Net模型, 双注意力模块, 编解码器, 特征补偿

Abstract:

Aiming at the problem that the background of high-resolution remote sensing images is complex and road extraction is easily disturbed by background information such as shadows, buildings, and railroads, the DAFCResUnet model with lightweight dual attention and feature compensation mechanism is proposed in this study. The model is based on ResUnet and achieves a balance between model performance and spatiotemporal complexity by adding lightweight dual attention and feature compensation modules. The dual attention module enhances the feature extraction capability of the model, and the feature compensation module can fuse the road features from deep and shallow layers in the network. The experimental results using DeepGlobe and GF-2 road datasets show that the IoU of the DAFCResUnet model can reach 0.6713, 0.8033, respectively, and the F1-score is 0.7402, 0.8507, respectively. The overall accuracy of the model is higher than that of U-Net, ResUnet, and VNet models. Compared with the U-Net and ResUnet models, the DAFCResUnet model only increases a small amount of computation and number of parameters, but the IoU and F1-score are improved substantially. Compared with the VNet model, the DAFCResUnet model achieves a higher accuracy with much lower computation and smaller number of parameters, and the model has advantages in both accuracy and spatiotemporal complexity. Compared with the other models, the DAFCResUnet model has stronger feature extraction and anti-interference ability, which can better solve the commission and omission caused by interfering objects on the road, ground features similar to roads, tree shade or shadow shading, etc.

Key words: deep learning, road extraction, high-resolution remote sensing images, residual network, U-Net model, dual attention module, codecs, feature compensation