HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection
Change detection (CD) from remote sensing images has been widely used in land management and urban planning. Benefiting from deep learning, numerous methods have achieved significant results in the CD of clearly changed targets. However, there are still significant challenges in the CD of weak targe...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10836868/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576739057860608 |
|---|---|
| author | Mingzhi Han Tao Xu Qingjie Liu Xiaohui Yang Jing Wang Jiaqi Kong |
| author_facet | Mingzhi Han Tao Xu Qingjie Liu Xiaohui Yang Jing Wang Jiaqi Kong |
| author_sort | Mingzhi Han |
| collection | DOAJ |
| description | Change detection (CD) from remote sensing images has been widely used in land management and urban planning. Benefiting from deep learning, numerous methods have achieved significant results in the CD of clearly changed targets. However, there are still significant challenges in the CD of weak targets, such as targets with small size, targets with blurred boundaries, and targets with low distinguishability from the background. Feature extraction from these targets can result in the loss of critical spatial features, potentially leading to decreased CD performance. Inspired by the improvement of multiscale features for CD of weak target, a hierarchical feature interaction network with multiscale fusion was proposed. First, a hierarchical feature interactive fusion module is proposed, which achieves optimized multichannel feature interaction and enhances the distinguishability between weak targets and background. Moreover, the module also achieves cross scale feature fusion, which compensates for the loss of spatial feature of changed targets at a single scale during feature extraction. Second, VMamba Block is utilized to obtain global features, and a spatial feature localization module was proposed to enhance the saliency of spatial features such as edges and textures. The distinguishability between weak targets and irrelevant spatial features is further enhanced. Our method has been experimentally evaluated on three public datasets, and outperformed state-of-the-art approaches by 1.06%, 1.41%, and 2.63% in F1 score on the LEVIR-CD, S2Looking, and NALand datasets, respectively. These results affirm the effectiveness of our method for weak targets in CD tasks. |
| format | Article |
| id | doaj-art-1a84ebdb425b497dbfb060d7568f3e5b |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-1a84ebdb425b497dbfb060d7568f3e5b2025-01-31T00:00:29ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01184318433010.1109/JSTARS.2025.352805310836868HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change DetectionMingzhi Han0https://orcid.org/0009-0003-1277-7781Tao Xu1https://orcid.org/0000-0002-4409-3777Qingjie Liu2https://orcid.org/0000-0002-5181-6451Xiaohui Yang3https://orcid.org/0000-0001-9677-979XJing Wang4Jiaqi Kong5https://orcid.org/0009-0002-5070-6901School of Information Science and Engineering, University of Jinan, Shandong, ChinaSchool of Information Science and Engineering, University of Jinan, Shandong, ChinaSchool of Computer Science and Engineering, Beihang University, Beijing, ChinaSchool of Information Science and Engineering, University of Jinan, Shandong, ChinaSchool of Information Science and Engineering, University of Jinan, Shandong, ChinaSchool of Information Science and Engineering, University of Jinan, Shandong, ChinaChange detection (CD) from remote sensing images has been widely used in land management and urban planning. Benefiting from deep learning, numerous methods have achieved significant results in the CD of clearly changed targets. However, there are still significant challenges in the CD of weak targets, such as targets with small size, targets with blurred boundaries, and targets with low distinguishability from the background. Feature extraction from these targets can result in the loss of critical spatial features, potentially leading to decreased CD performance. Inspired by the improvement of multiscale features for CD of weak target, a hierarchical feature interaction network with multiscale fusion was proposed. First, a hierarchical feature interactive fusion module is proposed, which achieves optimized multichannel feature interaction and enhances the distinguishability between weak targets and background. Moreover, the module also achieves cross scale feature fusion, which compensates for the loss of spatial feature of changed targets at a single scale during feature extraction. Second, VMamba Block is utilized to obtain global features, and a spatial feature localization module was proposed to enhance the saliency of spatial features such as edges and textures. The distinguishability between weak targets and irrelevant spatial features is further enhanced. Our method has been experimentally evaluated on three public datasets, and outperformed state-of-the-art approaches by 1.06%, 1.41%, and 2.63% in F1 score on the LEVIR-CD, S2Looking, and NALand datasets, respectively. These results affirm the effectiveness of our method for weak targets in CD tasks.https://ieeexplore.ieee.org/document/10836868/Change detection (CD)feature interactionmultiscale feature fusionremote sensing (RS) imageVMamba |
| spellingShingle | Mingzhi Han Tao Xu Qingjie Liu Xiaohui Yang Jing Wang Jiaqi Kong HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Change detection (CD) feature interaction multiscale feature fusion remote sensing (RS) image VMamba |
| title | HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection |
| title_full | HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection |
| title_fullStr | HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection |
| title_full_unstemmed | HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection |
| title_short | HFIFNet: Hierarchical Feature Interaction Network With Multiscale Fusion for Change Detection |
| title_sort | hfifnet hierarchical feature interaction network with multiscale fusion for change detection |
| topic | Change detection (CD) feature interaction multiscale feature fusion remote sensing (RS) image VMamba |
| url | https://ieeexplore.ieee.org/document/10836868/ |
| work_keys_str_mv | AT mingzhihan hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection AT taoxu hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection AT qingjieliu hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection AT xiaohuiyang hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection AT jingwang hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection AT jiaqikong hfifnethierarchicalfeatureinteractionnetworkwithmultiscalefusionforchangedetection |