DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data
Current machine learning methods for land subsidence susceptibility mapping (LSSM) predominantly focus on the spatial features of land subsidence conditioning factors (LSCFs), overlooking the sequence relationships that merger after the superposition of these factors. This often leads to unreliable...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-08-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2025.2499199 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849224338656460800 |
|---|---|
| author | Yi He Binghai Gao Haowen Yan Qing Zhang Lifeng Zhang Wende Li Xu He Jiangang Lu |
| author_facet | Yi He Binghai Gao Haowen Yan Qing Zhang Lifeng Zhang Wende Li Xu He Jiangang Lu |
| author_sort | Yi He |
| collection | DOAJ |
| description | Current machine learning methods for land subsidence susceptibility mapping (LSSM) predominantly focus on the spatial features of land subsidence conditioning factors (LSCFs), overlooking the sequence relationships that merger after the superposition of these factors. This often leads to unreliable LSSM results. To address this limitation, this paper proposes a novel double-branch parallel fusion neural network, termed DBPFNet, which integrates multi-factor sequence and spatial features to improve LSSM accuracy. The Beijing Plain is selected as the study area. InSAR-derived land subsidence data are used as positive samples, and 12 LSCFs are chosen for analysis. Convolutional neural network (CNN) is employed to learn multi-factor spatial features, while long short-term memory (LSTM) is used to learn multi-factor sequence features. The spatial and sequence features are fuzed by two full connections to generate the LSSM. Experimental results demonstrate that the proposed DBPFNet model significantly outperforms CNN, LSTM and transformer models in terms of performance, yielding highly accurate LSSM result. The high susceptibility areas are predominantly located in the central region of Beijing Plain. Key factors influencing land subsidence in the study area include groundwater, altitude, build density, precipitation and river density. |
| format | Article |
| id | doaj-art-0450f28edd554e9ca85e734e1adf9f5f |
| institution | Kabale University |
| issn | 1753-8947 1753-8955 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj-art-0450f28edd554e9ca85e734e1adf9f5f2025-08-25T11:24:45ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552025-08-0118110.1080/17538947.2025.2499199DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation dataYi He0Binghai Gao1Haowen Yan2Qing Zhang3Lifeng Zhang4Wende Li5Xu He6Jiangang Lu7Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaFaculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, People’s Republic of ChinaCurrent machine learning methods for land subsidence susceptibility mapping (LSSM) predominantly focus on the spatial features of land subsidence conditioning factors (LSCFs), overlooking the sequence relationships that merger after the superposition of these factors. This often leads to unreliable LSSM results. To address this limitation, this paper proposes a novel double-branch parallel fusion neural network, termed DBPFNet, which integrates multi-factor sequence and spatial features to improve LSSM accuracy. The Beijing Plain is selected as the study area. InSAR-derived land subsidence data are used as positive samples, and 12 LSCFs are chosen for analysis. Convolutional neural network (CNN) is employed to learn multi-factor spatial features, while long short-term memory (LSTM) is used to learn multi-factor sequence features. The spatial and sequence features are fuzed by two full connections to generate the LSSM. Experimental results demonstrate that the proposed DBPFNet model significantly outperforms CNN, LSTM and transformer models in terms of performance, yielding highly accurate LSSM result. The high susceptibility areas are predominantly located in the central region of Beijing Plain. Key factors influencing land subsidence in the study area include groundwater, altitude, build density, precipitation and river density.https://www.tandfonline.com/doi/10.1080/17538947.2025.2499199Land subsidence susceptibilityInSARCNNLSTMGIS and RS |
| spellingShingle | Yi He Binghai Gao Haowen Yan Qing Zhang Lifeng Zhang Wende Li Xu He Jiangang Lu DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data International Journal of Digital Earth Land subsidence susceptibility InSAR CNN LSTM GIS and RS |
| title | DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data |
| title_full | DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data |
| title_fullStr | DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data |
| title_full_unstemmed | DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data |
| title_short | DBPFNet: a double branch parallel fusion neural network method for land subsidence susceptibility mapping with InSAR observation data |
| title_sort | dbpfnet a double branch parallel fusion neural network method for land subsidence susceptibility mapping with insar observation data |
| topic | Land subsidence susceptibility InSAR CNN LSTM GIS and RS |
| url | https://www.tandfonline.com/doi/10.1080/17538947.2025.2499199 |
| work_keys_str_mv | AT yihe dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT binghaigao dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT haowenyan dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT qingzhang dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT lifengzhang dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT wendeli dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT xuhe dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata AT jianganglu dbpfnetadoublebranchparallelfusionneuralnetworkmethodforlandsubsidencesusceptibilitymappingwithinsarobservationdata |