Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery
Building extraction using very high resolution (VHR) optical remote sensing imagery is an essential interpretation task that impacts human life. However, buildings in different environments exhibit various scales, complicated spatial distributions, and different imaging conditions. Additionally, wit...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/7/1767 |
| _version_ | 1827622562695217152 |
|---|---|
| author | Jianhao Li Yin Zhuang Shan Dong Peng Gao Hao Dong He Chen Liang Chen Lianlin Li |
| author_facet | Jianhao Li Yin Zhuang Shan Dong Peng Gao Hao Dong He Chen Liang Chen Lianlin Li |
| author_sort | Jianhao Li |
| collection | DOAJ |
| description | Building extraction using very high resolution (VHR) optical remote sensing imagery is an essential interpretation task that impacts human life. However, buildings in different environments exhibit various scales, complicated spatial distributions, and different imaging conditions. Additionally, with the spatial resolution of images increasing, there are diverse interior details and redundant context information present in building and background areas. Thus, the above-mentioned situations would create large intra-class variances and poor inter-class discrimination, leading to uncertain feature descriptions for building extraction, which would result in over- or under-extraction phenomena. In this article, a novel hierarchical disentangling network with an encoder–decoder architecture called HDNet is proposed to consider both the stable and uncertain feature description in a convolution neural network (CNN). Next, a hierarchical disentangling strategy is set up to individually generate strong and weak semantic zones using a newly designed feature disentangling module (FDM). Here, the strong and weak semantic zones set up the stable and uncertain description individually to determine a more stable semantic main body and uncertain semantic boundary of buildings. Next, a dual-stream semantic feature description is built to gradually integrate strong and weak semantic zones by the designed component feature fusion module (CFFM), which is able to generate a powerful semantic description for more complete and refined building extraction. Finally, extensive experiments are carried out on three published datasets (i.e., WHU satellite, WHU aerial, and INRIA), and the comparison results show that the proposed HDNet outperforms other state-of-the-art (SOTA) methods. |
| first_indexed | 2024-03-09T11:27:59Z |
| format | Article |
| id | doaj.art-de93439faaf14b6c95f600db20d1bc8f |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T11:27:59Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-de93439faaf14b6c95f600db20d1bc8f2023-11-30T23:58:42ZengMDPI AGRemote Sensing2072-42922022-04-01147176710.3390/rs14071767Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing ImageryJianhao Li0Yin Zhuang1Shan Dong2Peng Gao3Hao Dong4He Chen5Liang Chen6Lianlin Li7Beijing Key Laboratory of Embedded Real-Time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, ChinaBeijing Key Laboratory of Embedded Real-Time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, ChinaBeijing Key Laboratory of Embedded Real-Time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, ChinaShanghai AI Laboratory, Shanghai 200232, ChinaCenter on Frontiers of Computing Studies and School of Electronic Engineering and Computer Science, Peking University, Beijing 100087, ChinaBeijing Key Laboratory of Embedded Real-Time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, ChinaBeijing Key Laboratory of Embedded Real-Time Information Processing Technology, Beijing Institute of Technology, Beijing 100081, ChinaCenter on Frontiers of Computing Studies and School of Electronic Engineering and Computer Science, Peking University, Beijing 100087, ChinaBuilding extraction using very high resolution (VHR) optical remote sensing imagery is an essential interpretation task that impacts human life. However, buildings in different environments exhibit various scales, complicated spatial distributions, and different imaging conditions. Additionally, with the spatial resolution of images increasing, there are diverse interior details and redundant context information present in building and background areas. Thus, the above-mentioned situations would create large intra-class variances and poor inter-class discrimination, leading to uncertain feature descriptions for building extraction, which would result in over- or under-extraction phenomena. In this article, a novel hierarchical disentangling network with an encoder–decoder architecture called HDNet is proposed to consider both the stable and uncertain feature description in a convolution neural network (CNN). Next, a hierarchical disentangling strategy is set up to individually generate strong and weak semantic zones using a newly designed feature disentangling module (FDM). Here, the strong and weak semantic zones set up the stable and uncertain description individually to determine a more stable semantic main body and uncertain semantic boundary of buildings. Next, a dual-stream semantic feature description is built to gradually integrate strong and weak semantic zones by the designed component feature fusion module (CFFM), which is able to generate a powerful semantic description for more complete and refined building extraction. Finally, extensive experiments are carried out on three published datasets (i.e., WHU satellite, WHU aerial, and INRIA), and the comparison results show that the proposed HDNet outperforms other state-of-the-art (SOTA) methods.https://www.mdpi.com/2072-4292/14/7/1767building extractionconvolution neural networksencoding–decoding methodhierarchical disentanglingoptical remote sensing imageryvery high resolution |
| spellingShingle | Jianhao Li Yin Zhuang Shan Dong Peng Gao Hao Dong He Chen Liang Chen Lianlin Li Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery Remote Sensing building extraction convolution neural networks encoding–decoding method hierarchical disentangling optical remote sensing imagery very high resolution |
| title | Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery |
| title_full | Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery |
| title_fullStr | Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery |
| title_full_unstemmed | Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery |
| title_short | Hierarchical Disentangling Network for Building Extraction from Very High Resolution Optical Remote Sensing Imagery |
| title_sort | hierarchical disentangling network for building extraction from very high resolution optical remote sensing imagery |
| topic | building extraction convolution neural networks encoding–decoding method hierarchical disentangling optical remote sensing imagery very high resolution |
| url | https://www.mdpi.com/2072-4292/14/7/1767 |
| work_keys_str_mv | AT jianhaoli hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT yinzhuang hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT shandong hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT penggao hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT haodong hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT hechen hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT liangchen hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery AT lianlinli hierarchicaldisentanglingnetworkforbuildingextractionfromveryhighresolutionopticalremotesensingimagery |