Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images
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 eff...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/14/2794 |
| _version_ | 1797526139052228608 |
|---|---|
| author | Shuhao Ran Xianjun Gao Yuanwei Yang Shaohua Li Guangbin Zhang Ping Wang |
| author_facet | Shuhao Ran Xianjun Gao Yuanwei Yang Shaohua Li Guangbin Zhang Ping Wang |
| author_sort | Shuhao Ran |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-10T09:24:58Z |
| format | Article |
| id | doaj.art-b0a985190eb849c7849483871159ae53 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T09:24:58Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-b0a985190eb849c7849483871159ae532023-11-22T04:52:30ZengMDPI AGRemote Sensing2072-42922021-07-011314279410.3390/rs13142794Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing ImagesShuhao Ran0Xianjun Gao1Yuanwei Yang2Shaohua Li3Guangbin Zhang4Ping Wang5School of Geosciences, Yangtze University, Wuhan 430100, ChinaSchool of Geosciences, Yangtze University, Wuhan 430100, ChinaSchool of Geosciences, Yangtze University, Wuhan 430100, ChinaSchool of Geosciences, Yangtze University, Wuhan 430100, ChinaSchool of Geosciences, Yangtze University, Wuhan 430100, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaDeep 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.https://www.mdpi.com/2072-4292/13/14/2794high-resolution remote sensing imagesbuilding extractionmultiscale featuresaggregate semantic informationfeature pyramid |
| spellingShingle | Shuhao Ran Xianjun Gao Yuanwei Yang Shaohua Li Guangbin Zhang Ping Wang Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images Remote Sensing high-resolution remote sensing images building extraction multiscale features aggregate semantic information feature pyramid |
| title | Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images |
| title_full | Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images |
| title_fullStr | Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images |
| title_full_unstemmed | Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images |
| title_short | Building Multi-Feature Fusion Refined Network for Building Extraction from High-Resolution Remote Sensing Images |
| title_sort | building multi feature fusion refined network for building extraction from high resolution remote sensing images |
| topic | high-resolution remote sensing images building extraction multiscale features aggregate semantic information feature pyramid |
| url | https://www.mdpi.com/2072-4292/13/14/2794 |
| work_keys_str_mv | AT shuhaoran buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages AT xianjungao buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages AT yuanweiyang buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages AT shaohuali buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages AT guangbinzhang buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages AT pingwang buildingmultifeaturefusionrefinednetworkforbuildingextractionfromhighresolutionremotesensingimages |