Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor
With the rapid development of urbanization, timely and accurate information on the spatial distribution of urban areas is essential for urban planning, environmental protection and sustainable urban development. To date, the main problem of urban mapping using synthetic aperture radar (SAR) data are...
| Glavni autori: | , , , , |
|---|---|
| Format: | Članak |
| Jezik: | English |
| Izdano: |
MDPI AG
2020-08-01
|
| Serija: | Remote Sensing |
| Teme: | |
| Online pristup: | https://www.mdpi.com/2072-4292/12/17/2791 |
| _version_ | 1827707397886443520 |
|---|---|
| author | Juanjuan Li Hong Zhang Chao Wang Fan Wu Lu Li |
| author_facet | Juanjuan Li Hong Zhang Chao Wang Fan Wu Lu Li |
| author_sort | Juanjuan Li |
| collection | DOAJ |
| description | With the rapid development of urbanization, timely and accurate information on the spatial distribution of urban areas is essential for urban planning, environmental protection and sustainable urban development. To date, the main problem of urban mapping using synthetic aperture radar (SAR) data are that nonbuilding objects with high backscattering cause high false alarms, while small-scale buildings with low backscattering result in omission errors. In this paper, a robust building-area extraction extractor is proposed to solve the above problems. The specific work includes (1) building a multiscale and multicategory building area dataset to learn enough building features in various areas; (2) designing a multiscale extraction network based on the residual convolutional block (ResNet50) and a pyramid-based pooling module to extract more discriminative features of building areas and introducing the focal loss item as the object function of the network to further extract the small-scale building areas and (3) eliminating the false alarms using the Normalized Difference Vegetation Index (NDVI) and Modified Normalized Difference Water Index (MNDWI) index. GF-3 SAR data with a 10-m resolution of four regions in China are used to validate our method, and the regional building-area mapping results with overall accuracy above 85% and kappa coefficient not less than 0.73 are obtained. Compared with the current popular networks and the global human settlement layer (GHSL) product, our method shows better extraction results and higher accuracy in multiscale building areas. The experiments using Sentinel-1 and ALOS-2/PALSAR-2 data show that the proposed method has good robustness with different SAR data sources. |
| first_indexed | 2024-03-10T16:46:04Z |
| format | Article |
| id | doaj.art-a81b53e753be4b9f90cd33ce6f7fd511 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T16:46:04Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a81b53e753be4b9f90cd33ce6f7fd5112023-11-20T11:38:35ZengMDPI AGRemote Sensing2072-42922020-08-011217279110.3390/rs12172791Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building ExtractorJuanjuan Li0Hong Zhang1Chao Wang2Fan Wu3Lu Li4Key Laboratory of Digital Earth Science, Aerospace information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaWith the rapid development of urbanization, timely and accurate information on the spatial distribution of urban areas is essential for urban planning, environmental protection and sustainable urban development. To date, the main problem of urban mapping using synthetic aperture radar (SAR) data are that nonbuilding objects with high backscattering cause high false alarms, while small-scale buildings with low backscattering result in omission errors. In this paper, a robust building-area extraction extractor is proposed to solve the above problems. The specific work includes (1) building a multiscale and multicategory building area dataset to learn enough building features in various areas; (2) designing a multiscale extraction network based on the residual convolutional block (ResNet50) and a pyramid-based pooling module to extract more discriminative features of building areas and introducing the focal loss item as the object function of the network to further extract the small-scale building areas and (3) eliminating the false alarms using the Normalized Difference Vegetation Index (NDVI) and Modified Normalized Difference Water Index (MNDWI) index. GF-3 SAR data with a 10-m resolution of four regions in China are used to validate our method, and the regional building-area mapping results with overall accuracy above 85% and kappa coefficient not less than 0.73 are obtained. Compared with the current popular networks and the global human settlement layer (GHSL) product, our method shows better extraction results and higher accuracy in multiscale building areas. The experiments using Sentinel-1 and ALOS-2/PALSAR-2 data show that the proposed method has good robustness with different SAR data sources.https://www.mdpi.com/2072-4292/12/17/2791building-area mappingdeep learningResNet50pyramid-pooling moduleGF-3 SAR |
| spellingShingle | Juanjuan Li Hong Zhang Chao Wang Fan Wu Lu Li Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor Remote Sensing building-area mapping deep learning ResNet50 pyramid-pooling module GF-3 SAR |
| title | Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor |
| title_full | Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor |
| title_fullStr | Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor |
| title_full_unstemmed | Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor |
| title_short | Spaceborne SAR Data for Regional Urban Mapping Using a Robust Building Extractor |
| title_sort | spaceborne sar data for regional urban mapping using a robust building extractor |
| topic | building-area mapping deep learning ResNet50 pyramid-pooling module GF-3 SAR |
| url | https://www.mdpi.com/2072-4292/12/17/2791 |
| work_keys_str_mv | AT juanjuanli spacebornesardataforregionalurbanmappingusingarobustbuildingextractor AT hongzhang spacebornesardataforregionalurbanmappingusingarobustbuildingextractor AT chaowang spacebornesardataforregionalurbanmappingusingarobustbuildingextractor AT fanwu spacebornesardataforregionalurbanmappingusingarobustbuildingextractor AT luli spacebornesardataforregionalurbanmappingusingarobustbuildingextractor |