A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution
High-resolution DEMs can provide accurate geographic information and can be widely used in hydrological analysis, path planning, and urban design. As the main complementary means of producing high-resolution DEMs, the DEM super-resolution (SR) method based on deep learning has reached a bottleneck....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/2/305 |
| _version_ | 1797437496291753984 |
|---|---|
| author | Xiaoyi Han Xiaochuan Ma Houpu Li Zhanlong Chen |
| author_facet | Xiaoyi Han Xiaochuan Ma Houpu Li Zhanlong Chen |
| author_sort | Xiaoyi Han |
| collection | DOAJ |
| description | High-resolution DEMs can provide accurate geographic information and can be widely used in hydrological analysis, path planning, and urban design. As the main complementary means of producing high-resolution DEMs, the DEM super-resolution (SR) method based on deep learning has reached a bottleneck. The reason for this phenomenon is that the DEM super-resolution method based on deep learning lacks a part of the global information it requires. Specifically, the multilevel aggregation process of deep learning has difficulty sufficiently capturing the low-level features with dependencies, which leads to a lack of global relationships with high-level information. To address this problem, we propose a global-information-constrained deep learning network for DEM SR (GISR). Specifically, our proposed GISR method consists of a global information supplement module and a local feature generation module. The former uses the Kriging method to supplement global information, considering the spatial autocorrelation rule. The latter includes a residual module and the PixelShuffle module, which is used to restore the detailed features of the terrain. Compared with the bicubic, Kriging, SRCNN, SRResNet, and TfaSR methods, the experimental results of our method show a better ability to retain terrain features, and the generation effect is more consistent with the ground truth DEM. Meanwhile, compared with the deep learning method, the RMSE of our results is improved by 20.5% to 68.8%. |
| first_indexed | 2024-03-09T11:21:10Z |
| format | Article |
| id | doaj.art-5c71abc0126e457895902ccbd04de85e |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T11:21:10Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-5c71abc0126e457895902ccbd04de85e2023-12-01T00:18:35ZengMDPI AGRemote Sensing2072-42922023-01-0115230510.3390/rs15020305A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-ResolutionXiaoyi Han0Xiaochuan Ma1Houpu Li2Zhanlong Chen3School of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaControl Engineering Laboratory, Naval University of Engineering, Wuhan 430030, ChinaSchool of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaHigh-resolution DEMs can provide accurate geographic information and can be widely used in hydrological analysis, path planning, and urban design. As the main complementary means of producing high-resolution DEMs, the DEM super-resolution (SR) method based on deep learning has reached a bottleneck. The reason for this phenomenon is that the DEM super-resolution method based on deep learning lacks a part of the global information it requires. Specifically, the multilevel aggregation process of deep learning has difficulty sufficiently capturing the low-level features with dependencies, which leads to a lack of global relationships with high-level information. To address this problem, we propose a global-information-constrained deep learning network for DEM SR (GISR). Specifically, our proposed GISR method consists of a global information supplement module and a local feature generation module. The former uses the Kriging method to supplement global information, considering the spatial autocorrelation rule. The latter includes a residual module and the PixelShuffle module, which is used to restore the detailed features of the terrain. Compared with the bicubic, Kriging, SRCNN, SRResNet, and TfaSR methods, the experimental results of our method show a better ability to retain terrain features, and the generation effect is more consistent with the ground truth DEM. Meanwhile, compared with the deep learning method, the RMSE of our results is improved by 20.5% to 68.8%.https://www.mdpi.com/2072-4292/15/2/305DEM super-resolutionspatial autocorrelationKrigingResNet |
| spellingShingle | Xiaoyi Han Xiaochuan Ma Houpu Li Zhanlong Chen A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution Remote Sensing DEM super-resolution spatial autocorrelation Kriging ResNet |
| title | A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution |
| title_full | A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution |
| title_fullStr | A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution |
| title_full_unstemmed | A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution |
| title_short | A Global-Information-Constrained Deep Learning Network for Digital Elevation Model Super-Resolution |
| title_sort | global information constrained deep learning network for digital elevation model super resolution |
| topic | DEM super-resolution spatial autocorrelation Kriging ResNet |
| url | https://www.mdpi.com/2072-4292/15/2/305 |
| work_keys_str_mv | AT xiaoyihan aglobalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT xiaochuanma aglobalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT houpuli aglobalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT zhanlongchen aglobalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT xiaoyihan globalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT xiaochuanma globalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT houpuli globalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution AT zhanlongchen globalinformationconstraineddeeplearningnetworkfordigitalelevationmodelsuperresolution |