Attention Enhanced U-Net for Building Extraction from Farmland Based on Google and WorldView-2 Remote Sensing Images

被引:0
|
作者
Li, Chuangnong [1 ]
Fu, Lin [1 ]
Zhu, Qing [1 ]
Zhu, Jun [1 ]
Fang, Zheng [2 ]
Xie, Yakun [1 ]
Guo, Yukun [1 ]
Gong, Yuhang [2 ]
机构
[1] Southwest Jiaotong Univ, Fac Geosci & Environm Engn, Chengdu 611756, Peoples R China
[2] Sichuan Inst Land Sci & Technol, Sichuan Ctr Satellite Applicat Technol, Chengdu 610041, Peoples R China
关键词
building extraction; farmland range; attention enhancement; U-Net network improvement; multi-source remote sensing image; NETWORK;
D O I
10.3390/rs13214411
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
High-resolution remote sensing images contain abundant building information and provide an important data source for extracting buildings, which is of great significance to farmland preservation. However, the types of ground features in farmland are complex, the buildings are scattered and may be obscured by clouds or vegetation, leading to problems such as a low extraction accuracy in the existing methods. In response to the above problems, this paper proposes a method of attention-enhanced U-Net for building extraction from farmland, based on Google and WorldView-2 remote sensing images. First, a Resnet unit is adopted as the infrastructure of the U-Net network encoding part, then the spatial and channel attention mechanism module is introduced between the Resnet unit and the maximum pool and the multi-scale fusion module is added to improve the U-Net network. Second, the buildings found on WorldView-2 and Google images are extracted through farmland boundary constraints. Third, boundary optimization and fusion processing are carried out for the building extraction results on the WorldView-2 and Google images. Fourth, a case experiment is performed. The method in this paper is compared with semantic segmentation models, such as FCN8, U-Net, Attention_UNet, and DeepLabv3+. The experimental results indicate that this method attains a higher accuracy and better effect in terms of building extraction within farmland; the accuracy is 97.47%, the F1 score is 85.61%, the recall rate (Recall) is 93.02%, and the intersection of union (IoU) value is 74.85%. Hence, buildings within farming areas can be effectively extracted, which is conducive to the preservation of farmland.
引用
收藏
页数:15
相关论文
共 50 条
  • [1] Building Extraction from Remote Sensing Images Based on Improved U-Net
    Jin Shu
    Guan Mo
    Bian Yuchan
    Wang Shulei
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2023, 60 (04)
  • [2] Building extraction from remote sensing images using deep residual U-Net
    Wang, Haiying
    Miao, Fang
    [J]. EUROPEAN JOURNAL OF REMOTE SENSING, 2022, 55 (01) : 71 - 85
  • [3] Building Detection from Remote Sensing Images Based on Improved U-net
    Ren Xinlei
    Wang Yangping
    Yang Jingyu
    Gao Decheng
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2019, 56 (22)
  • [4] The extraction of the desert roads based on the U-Net network from remote sensing images
    Zhang, Xubing
    Li, Can
    Shao, Shiwei
    Peng, Yuxin
    [J]. INTERNATIONAL CONFERENCE ON ENVIRONMENTAL REMOTE SENSING AND BIG DATA (ERSBD 2021), 2021, 12129
  • [5] Dual Hybrid Attention Mechanism-Based U-Net for Building Segmentation in Remote Sensing Images
    Lei, Jingxiong
    Liu, Xuzhi
    Yang, Haolang
    Zeng, Zeyu
    Feng, Jun
    [J]. APPLIED SCIENCES-BASEL, 2024, 14 (03):
  • [6] Rural Building Extraction Based on Joint U-Net and the Generalized Chinese Restaurant Franchise from Remote Sensing Images
    Wang, Zixiong
    Li, Shaodan
    Zhu, Zimeng
    [J]. SUSTAINABILITY, 2023, 15 (05)
  • [7] Convolutional block attention module U-Net: a method to improve attention mechanism and U-Net for remote sensing images
    Zhang, Yanjun
    Kong, Jiayuan
    Long, Sifang
    Zhu, Yuanhao
    He, Fushuai
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2022, 16 (02)
  • [8] A landslide extraction method of channel attention mechanism U-Net network based on Sentinel-2A remote sensing images
    Chen, Hesheng
    He, Yi
    Zhang, Lifeng
    Yao, Sheng
    Yang, Wang
    Fang, Yumin
    Liu, Yaoxiang
    Gao, Binghai
    [J]. INTERNATIONAL JOURNAL OF DIGITAL EARTH, 2023, 16 (01) : 552 - 577
  • [9] AN IMPROVED U-NET MODEL FOR BUILDINGS EXTRACTION WITH REMOTE SENSING IMAGES
    He, Weibing
    Qiang, Xiaoyong
    Maihaimaiti, Azigu
    Chen, Shengyi
    Ge, Bingfu
    Huang, Fang
    [J]. IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 6870 - 6873
  • [10] Reconstruction Bias U-Net for Road Extraction From Optical Remote Sensing Images
    Chen, Ziyi
    Wang, Cheng
    Li, Jonathan
    Xie, Nianci
    Han, Yan
    Du, Jixiang
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2021, 14 : 2284 - 2294