A cross-stage features fusion network for building extraction from remote sensing images
The deep learning-based building extraction methods produce different feature maps at different stages of the network, which contain different information features. The detailed information of the feature maps decreases along the depth of the network, and insufficiently detailed information results...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-03-01
|
| Series: | Geo-spatial Information Science |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10095020.2024.2307922 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839649773834993664 |
|---|---|
| author | Xiaolong Zuo Zhenfeng Shao Jiaming Wang Xiao Huang Yu Wang |
| author_facet | Xiaolong Zuo Zhenfeng Shao Jiaming Wang Xiao Huang Yu Wang |
| author_sort | Xiaolong Zuo |
| collection | DOAJ |
| description | The deep learning-based building extraction methods produce different feature maps at different stages of the network, which contain different information features. The detailed information of the feature maps decreases along the depth of the network, and insufficiently detailed information results in limited accuracy. However, existing methods are incapable of making full use of low-level feature maps with rich details. To overcome these shortcomings, we proposed a Cross-stage Features Fusion Network (CFF-Net) for building extraction from remote sensing images. In the CFF-Net, we innovatively proposed a Cross-stage Features Fusion (CFF) module that fuses different features generated at different stages. And we used the attention mechanism to make the network more focused on important information at different scales. To further improve the accuracy of building extraction, we designed the Prediction Enhancement (PE) module, where the last convolutional layer and the feature map generated in the intermediate stage are used for prediction at the same time to enhance the final result. To evaluate the effectiveness of the proposed network, we conduct quantitative and qualitative experiments on the two publicly available datasets, i.e. the Inria dataset and the WHU datasets. CFF-Net outperformed other state-of-the-art algorithms on the two datasets in IoU and F1 metrics. The efficiency analysis reveals that the proposed CFF-Net achieves a great balance between building extraction performance and complexity/efficiency, with faster convergence and higher robustness. |
| format | Article |
| id | doaj-art-b4d4272b14b44f06b485e7de067be1e5 |
| institution | Matheson Library |
| issn | 1009-5020 1993-5153 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geo-spatial Information Science |
| spelling | doaj-art-b4d4272b14b44f06b485e7de067be1e52025-06-27T09:55:27ZengTaylor & Francis GroupGeo-spatial Information Science1009-50201993-51532025-03-0128238740110.1080/10095020.2024.2307922A cross-stage features fusion network for building extraction from remote sensing imagesXiaolong Zuo0Zhenfeng Shao1Jiaming Wang2Xiao Huang3Yu Wang4State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaHubei Key Laboratory of Intelligent Robot, Wuhan Institute of Technology, Wuhan, ChinaDepartment of Environmental Sciences, Emory University, Atlanta, GA, USAState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaThe deep learning-based building extraction methods produce different feature maps at different stages of the network, which contain different information features. The detailed information of the feature maps decreases along the depth of the network, and insufficiently detailed information results in limited accuracy. However, existing methods are incapable of making full use of low-level feature maps with rich details. To overcome these shortcomings, we proposed a Cross-stage Features Fusion Network (CFF-Net) for building extraction from remote sensing images. In the CFF-Net, we innovatively proposed a Cross-stage Features Fusion (CFF) module that fuses different features generated at different stages. And we used the attention mechanism to make the network more focused on important information at different scales. To further improve the accuracy of building extraction, we designed the Prediction Enhancement (PE) module, where the last convolutional layer and the feature map generated in the intermediate stage are used for prediction at the same time to enhance the final result. To evaluate the effectiveness of the proposed network, we conduct quantitative and qualitative experiments on the two publicly available datasets, i.e. the Inria dataset and the WHU datasets. CFF-Net outperformed other state-of-the-art algorithms on the two datasets in IoU and F1 metrics. The efficiency analysis reveals that the proposed CFF-Net achieves a great balance between building extraction performance and complexity/efficiency, with faster convergence and higher robustness.https://www.tandfonline.com/doi/10.1080/10095020.2024.2307922Remote sensingcross-stage features fusionbuilding extractionattention mechanism |
| spellingShingle | Xiaolong Zuo Zhenfeng Shao Jiaming Wang Xiao Huang Yu Wang A cross-stage features fusion network for building extraction from remote sensing images Geo-spatial Information Science Remote sensing cross-stage features fusion building extraction attention mechanism |
| title | A cross-stage features fusion network for building extraction from remote sensing images |
| title_full | A cross-stage features fusion network for building extraction from remote sensing images |
| title_fullStr | A cross-stage features fusion network for building extraction from remote sensing images |
| title_full_unstemmed | A cross-stage features fusion network for building extraction from remote sensing images |
| title_short | A cross-stage features fusion network for building extraction from remote sensing images |
| title_sort | cross stage features fusion network for building extraction from remote sensing images |
| topic | Remote sensing cross-stage features fusion building extraction attention mechanism |
| url | https://www.tandfonline.com/doi/10.1080/10095020.2024.2307922 |
| work_keys_str_mv | AT xiaolongzuo acrossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT zhenfengshao acrossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT jiamingwang acrossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT xiaohuang acrossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT yuwang acrossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT xiaolongzuo crossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT zhenfengshao crossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT jiamingwang crossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT xiaohuang crossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages AT yuwang crossstagefeaturesfusionnetworkforbuildingextractionfromremotesensingimages |