Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images

被引:23
|
作者
Ran, Shuhao [1 ]
Gao, Xianjun [1 ]
Yang, Yuanwei [1 ,2 ,3 ]
Li, Shaohua [1 ]
Zhang, Guangbin [1 ]
Wang, Ping [4 ,5 ]
机构
[1] Yangtze Univ, Sch Geosci, Wuhan 430100, Peoples R China
[2] Beijing Key Lab Urban Spatial Informat Engn, Beijing 100045, Peoples R China
[3] Hunan Univ Sci & Technol, Hunan Prov Key Lab Geoinformat Engn Surveying Map, Xiangtan 411201, Peoples R China
[4] Chinese Acad Sci, Aerosp Informat Res Inst, Beijing 100094, Peoples R China
[5] Key Lab Earth Observat Hainan Prov, Sanya 572029, Peoples R China
基金
中国国家自然科学基金;
关键词
high-resolution remote sensing images; building extraction; multiscale features; aggregate semantic information; feature pyramid; SEMANTIC SEGMENTATION; AERIAL IMAGES; CLASSIFICATION;
D O I
10.3390/rs13142794
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Deep learning approaches have been widely used in building automatic extraction tasks and have made great progress in recent years. However, the missing detection and wrong detection causing by spectrum confusion is still a great challenge. The existing fully convolutional networks (FCNs) cannot effectively distinguish whether the feature differences are from one building or the building and its adjacent non-building objects. In order to overcome the limitations, a building multi-feature fusion refined network (BMFR-Net) was presented in this paper to extract buildings accurately and completely. BMFR-Net is based on an encoding and decoding structure, mainly consisting of two parts: the continuous atrous convolution pyramid (CACP) module and the multiscale output fusion constraint (MOFC) structure. The CACP module is positioned at the end of the contracting path and it effectively minimizes the loss of effective information in multiscale feature extraction and fusion by using parallel continuous small-scale atrous convolution. To improve the ability to aggregate semantic information from the context, the MOFC structure performs predictive output at each stage of the expanding path and integrates the results into the network. Furthermore, the multilevel joint weighted loss function effectively updates parameters well away from the output layer, enhancing the learning capacity of the network for low-level abstract features. The experimental results demonstrate that the proposed BMFR-Net outperforms the other five state-of-the-art approaches in both visual interpretation and quantitative evaluation.
引用
收藏
页数:24
相关论文
共 50 条
  • [1] Multi-Scale Attention Network for Building Extraction from High-Resolution Remote Sensing Images
    Chang, Jing
    He, Xiaohui
    Li, Panle
    Tian, Ting
    Cheng, Xijie
    Qiao, Mengjia
    Zhou, Tao
    Zhang, Beibei
    Chang, Ziqian
    Fan, Tingwei
    SENSORS, 2024, 24 (03)
  • [2] Hierarchical Building Extraction from High-resolution Remote Sensing Imagery Based on Multi-feature and Multi-scale Method
    Lin, Yuzhun
    Zhang, Baoming
    Xu, Junfeng
    Li, Jian
    Zhao, Chuan
    Yu, Donghang
    PROCEEDINGS OF 2018 THE 3RD INTERNATIONAL CONFERENCE ON MULTIMEDIA AND IMAGE PROCESSING (ICMIP 2018), 2018, : 17 - 23
  • [3] Depth Feature Fusion Network for Building Extraction in Remote Sensing Images
    Chen, Junming
    Liu, Bing
    Yu, Anzhu
    Quan, Yujun
    Li, Tingting
    Guo, Wenyue
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2024, 17 : 16577 - 16591
  • [4] Multi-Level Perceptual Network for Urban Building Extraction from High-Resolution Remote Sensing Images
    Sun, Yueming
    Chen, Jinlong
    Huang, Xiao
    Zhang, Hongsheng
    PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING, 2023, 89 (07): : 427 - 434
  • [5] A Context Feature Enhancement Network for Building Extraction from High-Resolution Remote Sensing Imagery
    Chen, Jinzhi
    Zhang, Dejun
    Wu, Yiqi
    Chen, Yilin
    Yan, Xiaohu
    REMOTE SENSING, 2022, 14 (09)
  • [6] BUILDING EXTRACTION FROM HIGH-RESOLUTION REMOTE SENSING IMAGERY BASED ON MULTI-SCALE FEATURE FUSION AND ENHANCEMENT
    Chen, Y.
    Cheng, H.
    Yao, S.
    Hu, Z.
    XXIV ISPRS CONGRESS: IMAGING TODAY, FORESEEING TOMORROW, COMMISSION III, 2022, 43-B3 : 55 - 60
  • [7] BEARNet: A Novel Buildings Edge-Aware Refined Network for Building Extraction From High-Resolution Remote Sensing Images
    Lin, Huijing
    Hao, Ming
    Luo, Weiqiang
    Yu, Hongye
    Zheng, Nanshan
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2023, 20
  • [8] Multi-Scale Feature Fusion Attention Network for Building Extraction in Remote Sensing Images
    Liu, Jia
    Gu, Hang
    Li, Zuhe
    Chen, Hongyang
    Chen, Hao
    ELECTRONICS, 2024, 13 (05)
  • [9] A Multimodal Feature Fusion Network for Building Extraction With Very High-Resolution Remote Sensing Image and LiDAR Data
    Luo, Hui
    Feng, Xibo
    Du, Bo
    Zhang, Yuxiang
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62 : 1 - 19
  • [10] Deep Learning for Building Extraction from High-Resolution Remote Sensing Images
    Norelyaqine, Abderrahim
    Saadane, Abderrahim
    ADVANCED TECHNOLOGIES FOR HUMANITY, 2022, 110 : 116 - 128