标題: A multi-terrain feature-based deep convolutional neural network for constructing super-resolution DEMs

作者: Zhou, AN (Zhou, Annan); Chen, YM (Chen, Yumin); Wilson, JP (Wilson, John P.); Chen, GD (Chen, Guodong); Min, WK (Min, Wankun); Xu, R (Xu, Rui)

來源出版物: INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 卷: 120 文獻号: 103338 DOI: 10.1016/j.jag.2023.103338 提前訪問日期: MAY 2023 出版年: JUN 2023

摘要: Scale conversion between DEMs is an important issue in geomorphometry. There are many mature studies on the generation of low-resolution(LR) DEMs from high-resolution(HR) DEMs. However, as an important and conve-nient means of obtaining HR DEMs, traditional super resolution (SR) methods have shown insufficient consid-eration of the terrain features embedded in DEMs. Therefore, this article investigates the combination of terrain features and the use of convolutional neural networks (CNN) to reconstruct HR DEMs, and proposes a multi -terrain feature-based deep CNN for super-resolution(SR) DEMs (MTF-SR). In the experiments, from the perspective of vector and raster terrain features, we fuse raster terrain features in the input and loss functions, and fuse vector terrain features in the optimization of the output of the model. The results show that the MTF-SR model has a 30-50 % reduction in mean absolute error (MAE) compared with interpolation methods, has the lowest slope and aspect error and has a 10 to 30 % improvement in streamline matching rate (SMR). These results point to the advantages of the method in overall elevation accuracy and the preservation of terrain features.

作者關鍵詞: Terrain features; Convolutional neural networks; Digital elevation models; Super -resolution

地址: [Zhou, Annan; Chen, Yumin; Chen, Guodong; Min, Wankun; Xu, Rui] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Hubei, Peoples R China.

[Chen, Yumin] Wuhan Univ, Key Lab Geog Informat Syst, Minist Educ, 129 Luoyu Rd, Wuhan 430079, Peoples R China.

[Chen, Yumin] Wuhan Univ, Key Lab Digital Cartog & Land Informat Applicat, Minist Nat Resources Peoples Republ China, 129 Luoyu Rd, Wuhan 430079, Peoples R China.

[Wilson, John P.] Univ Southern Calif Los Angeles, Spatial Sci Inst, Los Angeles, CA 90089 USA.

[Chen, Yumin] 129 Luoyu Rd, Wuhan, Hubei, Peoples R China.

通訊作者地址: Chen, YM (通訊作者)，129 Luoyu Rd, Wuhan, Hubei, Peoples R China.

電子郵件地址: zhouannan2016@whu.edu.cn; ymchen@whu.edu.cn; jpwilson@usc.edu; guodongchen@whu.edu.cn; minwankun@whu.edu.cn; 2022202050059@whu.edu.cn

影響因子：7.5