RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks
The digital elevation model (DEM) is known as one kind of the most significant fundamental geographical data models. The theory, method and application of DEM are hot research issues in geography, especially in geomorphology, hydrology, soil and other related fields. In this paper, we improve the ef...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/10/8/501 |
| _version_ | 1827685092506468352 |
|---|---|
| author | Ruichen Zhang Shaofeng Bian Houpu Li |
| author_facet | Ruichen Zhang Shaofeng Bian Houpu Li |
| author_sort | Ruichen Zhang |
| collection | DOAJ |
| description | The digital elevation model (DEM) is known as one kind of the most significant fundamental geographical data models. The theory, method and application of DEM are hot research issues in geography, especially in geomorphology, hydrology, soil and other related fields. In this paper, we improve the efficient sub-pixel convolutional neural networks (ESPCN) and propose recursive sub-pixel convolutional neural networks (RSPCN) to generate higher-resolution DEMs (HRDEMs) from low-resolution DEMs (LRDEMs). Firstly, the structure of RSPCN is described in detail based on recursion theory. This paper explores the effects of different training datasets, with the self-adaptive learning rate Adam algorithm optimizing the model. Furthermore, the adding-“zero” boundary method is introduced into the RSPCN algorithm as a data preprocessing method, which improves the RSPCN method’s accuracy and convergence. Extensive experiments are conducted to train the method till optimality. Finally, comparisons are made with other traditional interpolation methods, such as bicubic, nearest-neighbor and bilinear methods. The results show that our method has obvious improvements in both accuracy and robustness and further illustrate the feasibility of deep learning methods in the DEM data processing area. |
| first_indexed | 2024-03-10T08:46:07Z |
| format | Article |
| id | doaj.art-a905d8afc7eb40f7a508779427d4a4f8 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-10T08:46:07Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-a905d8afc7eb40f7a508779427d4a4f82023-11-22T07:52:50ZengMDPI AGISPRS International Journal of Geo-Information2220-99642021-07-0110850110.3390/ijgi10080501RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural NetworksRuichen Zhang0Shaofeng Bian1Houpu Li2Department of Electrical Engineering, Naval University of Engineering, Wuhan 430033, ChinaDepartment of Electrical Engineering, Naval University of Engineering, Wuhan 430033, ChinaDepartment of Electrical Engineering, Naval University of Engineering, Wuhan 430033, ChinaThe digital elevation model (DEM) is known as one kind of the most significant fundamental geographical data models. The theory, method and application of DEM are hot research issues in geography, especially in geomorphology, hydrology, soil and other related fields. In this paper, we improve the efficient sub-pixel convolutional neural networks (ESPCN) and propose recursive sub-pixel convolutional neural networks (RSPCN) to generate higher-resolution DEMs (HRDEMs) from low-resolution DEMs (LRDEMs). Firstly, the structure of RSPCN is described in detail based on recursion theory. This paper explores the effects of different training datasets, with the self-adaptive learning rate Adam algorithm optimizing the model. Furthermore, the adding-“zero” boundary method is introduced into the RSPCN algorithm as a data preprocessing method, which improves the RSPCN method’s accuracy and convergence. Extensive experiments are conducted to train the method till optimality. Finally, comparisons are made with other traditional interpolation methods, such as bicubic, nearest-neighbor and bilinear methods. The results show that our method has obvious improvements in both accuracy and robustness and further illustrate the feasibility of deep learning methods in the DEM data processing area.https://www.mdpi.com/2220-9964/10/8/501DEMsuper-resolutioninterpolationrecursiondeep learning |
| spellingShingle | Ruichen Zhang Shaofeng Bian Houpu Li RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks ISPRS International Journal of Geo-Information DEM super-resolution interpolation recursion deep learning |
| title | RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks |
| title_full | RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks |
| title_fullStr | RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks |
| title_full_unstemmed | RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks |
| title_short | RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks |
| title_sort | rspcn super resolution of digital elevation model based on recursive sub pixel convolutional neural networks |
| topic | DEM super-resolution interpolation recursion deep learning |
| url | https://www.mdpi.com/2220-9964/10/8/501 |
| work_keys_str_mv | AT ruichenzhang rspcnsuperresolutionofdigitalelevationmodelbasedonrecursivesubpixelconvolutionalneuralnetworks AT shaofengbian rspcnsuperresolutionofdigitalelevationmodelbasedonrecursivesubpixelconvolutionalneuralnetworks AT houpuli rspcnsuperresolutionofdigitalelevationmodelbasedonrecursivesubpixelconvolutionalneuralnetworks |