-
摘要: 融合多源数字高程模型(digital elevation model, DEM)数据能有效利用数据间的互补优势，提升单一源数据的质量。提出一种基于深度置信网络(deep belief networks, DBN)的点面融合模型，在DBN的框架下考虑地形坡度、地表覆盖和空间位置信息等因素对DEM高程误差空间分布的影响，建立DEM高程值与高精度激光雷达测高数据之间的回归关系，从而实现多源栅格DEM与激光雷达测高点数据的空间融合，提升栅格DEM的垂直精度。对于空洞数据，根据空洞和非空洞区域的范围建立相应的输入数据集，分别进行融合，再利用不规则三角网差分曲面方法实现融合结果的无缝拼接。实验结果表明，相比原始DEM数据和两两融合的结果，所提出的多源DEM点面融合模型能够大幅度提升数据精度，有效解决原始数据中存在的空洞、噪声和异常值等问题。Abstract:Objectives Although there have been a variety of global digital elevation model (DEM) products available for geospatial applications, the data quality of DEMs are inevitably affected by the observing techniques and processing methods. Multi-source data fusion by integrating the complementary information among the datasets is an effective way to improve DEM data quality. Based on the facts, this paper proposed a point-surface fusion model based on deep belief network (DBN).Methods The terrain slope, land cover, and the spatial coordinates were integrated into the DBN model, thus constructed the statistical relation between the DEM elevations and the high-quality ICESat (ice, cloud, and land elevation) GLAS (geoscience laser altimeter system) data. The DBN-based fusion model was capable to learn the spatial structure and landscape associations of the DEM errors, and thus the multi-source raster DEMs and the high-quality LiDAR altimetry can be effectively fused. For the data tiles with voids, the triangulated irregular network (TIN) delta surface method was employed to merge the fusion results for the void and non-void areas, respectively.Results In the experiments, the effectiveness of the proposed method is tested on the fusion of ASTER (advanced spaceborne thermal emission and reflection radiometer) GDEM (global digital elevation model) v2, ALOS AW3D30 (advanced land observing satellite world 3D-30 m) and SRTM1 (shuttle radar topography mission 1-arc second) data, which are the most popular DEM products. Six data tiles with different terrain conditions are tested, and the accuracy of the original data and the fusion results are evaluated using the ICESat GLAS points with cross-validation. Quantitative results show that the fused data have higher accuracy compared with the original datasets and the pair-wise fusion results, both in terms of the void areas and the non-void areas.Conclusions The results show that the proposed method is able to improve the accuracy of the DEM data with fusion of multi-source altimetry datasets. Moreover, it is effective in dealing with the data voids, noise and abnormal values simultaneously.
-
-
![]()
图 1 DBN网络及点面融合模型的基本结构
Figure 1. Basic Structures of DBN and the Point-Surface Fusion Model
![]()
图 3 N29E094数据块局部区域的视觉比较
Figure 3. Visual Comparison of a Local Region Within the Tile N29E094
![]()
图 4 N24E107和S02W079数据块局部区域的视觉比较
Figure 4. Visual Comparison of Two Local Areas Within Tiles N24E107 and S02W079, Respectively
表 1 交叉验证实验定量评价结果(RMSE值)/m
Table 1 Quantitative Evaluation Results (RMSE) in the Cross-Validation Experiments/m
数据块 训练精度 验证精度 ASTER GDEM AW3D30 SRTM1 本文方法 ASTER GDEM AW3D30 SRTM1 本文方法 N28E101 39.289 20.836 21.751 17.453 39.223 20.788 21.712 17.696 N31E102 27.844 15.249 15.927 12.536 27.836 15.236 15.915 12.976 S02W079 35.597 13.305 12.732 11.251 32.802 13.275 12.696 11.267 N24E107 21.823 16.178 15.448 13.204 21.789 16.149 15.425 13.422 N29E094 33.319 58.797 20.332 15.778 33.271 55.415 20.300 15.879 N35W102 7.297 5.946 6.259 1.538 7.293 5.945 6.259 1.524 
下载: 导出CSV
表 2 无空洞数据整体定量评价结果/m
Table 2 Quantitative Evaluation Results for the Non-void Data/m
数据块 精度指标 ASTER GDEM AW3D30 SRTM1 ASTER_c AW3D30_c SRTM1_c 本文方法 N29E094 MEAN -14.839 -12.266 -9.717 0.404 -0.192 0.120 0.067 RMSE 33.526 58.839 20.824 26.570 49.951 16.930 15.293 N35W102 MEAN -0.671 -5.323 -5.138 -0.001 0.017 0.037 -0.001 RMSE 7.310 5.955 6.269 4.371 1.515 2.196 1.403
下载: 导出CSV
表 3 无空洞区域和空洞区域的定量评价结果/m
Table 3 Quantitative Evaluation Results for the Non-void Areas and Void Areas/m
数据块 精度指标 无空洞区域 空洞区域 ASTER GDEM AW3D30 SRTM1 本文方法 ASTER GDEM AW3D30 BP_TIN 本文方法 N28E101 MEAN -12.499 -5.285 -4.358 -0.359 -27.495 -10.231 -8.347 0.069 RMSE 29.002 14.358 14.182 11.497 55.009 30.263 46.570 25.942 N31E102 MEAN -14.582 -3.682 -3.230 -0.206 -21.367 -14.094 -6.205 -4.139 RMSE 27.083 14.619 13.445 12.006 48.979 31.070 36.276 28.151 S02W079 MEAN 3.509 -2.147 -1.737 0.286 -51.750 -3.121 -15.648 -1.822 RMSE 23.694 12.767 10.557 10.319 206.618 31.581 66.973 30.112 N24E107 MEAN -0.940 -5.905 -3.439 0.306 -8.773 -7.888 -4.708 0.734 RMSE 18.397 13.921 12.808 11.262 30.901 22.319 28.320 18.544
下载: 导出CSV
-
[1] 汤国安. 我国数字高程模型与数字地形分析研究进展[J]. 地理学报, 2014, 69(9): 1 305-1 325 https://www.cnki.com.cn/Article/CJFDTOTAL-DLXB201409007.htm Tang Guoan. Progress of DEM and Digital Terrain Analysis in China[J]. Acta Geographica Sinica, 2014, 69(9): 1 305-1 325 https://www.cnki.com.cn/Article/CJFDTOTAL-DLXB201409007.htm
[2] Yang Liping, Meng Xingmin, Zhang Xiaoqiang. SRTM DEM and Its Application Advances[J]. International Journal of Remote Sensing, 2011, 32(14): 3 875-3 896 doi: 10.1080/01431161003786016
[3] Tachikawa T, Hato M, Kaku M, et al. Characteristics of ASTER GDEM Version 2[C]. 2011 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Vancouver, BC, Canada, 2011
[4] Takaku J, Tadono T. Quality Updates of 'AW3D' Global DSM Generated from ALOS PRISM[C]. 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Fort Worth, TX, USA, 2017
[5] 陈楠. DEM分辨率变化对坡度误差的影响[J]. 武汉大学学报·信息科学版, 2013, 38(5): 594-598 http://ch.whu.edu.cn/article/id/2650 Chen Nan. Influence of Resolutions of DEM on the Error of Slope[J]. Geomatics and Information Science of Wuhan University, 2013, 38(5): 594-598 http://ch.whu.edu.cn/article/id/2650
[6] Toutin T. Impact of Terrain Slope and Aspect on Radargrammetric DEM Accuracy[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2002, 57(3): 228-240 doi: 10.1016/S0924-2716(02)00123-5
[7] 李振洪, 李鹏, 丁咚, 等. 全球高分辨率数字高程模型研究进展与展望[J]. 武汉大学学报·信息科学版, 2018, 43(12): 1 927-1 942 doi: 10.13203/j.whugis20180295 Li Zhenhong, Li Peng, Ding Dong, et al. Research Progress of Global High Resolution Digital Elevation Models[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 1 927-1 942 doi: 10.13203/j.whugis20180295
[8] Rizzoli P, Martone M, Gonzalez C, et al. Generation and Performance Assessment of the Global TanDEM-X Digital Elevation Model[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 132: 119-139 doi: 10.1016/j.isprsjprs.2017.08.008
[9] Takaku J, Tadono T, Tsutsui K, et al. Quality Improvements of AW3D Global DSM Derived from ALOS PRISM[C]. 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 2018
[10] 唐新明, 王鸿燕, 祝小勇. 资源三号卫星测绘技术与应用[J]. 测绘学报, 2017, 46(10): 1 482-1 491 https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201710031.htm Tang Xinming, Wang Hongyan, Zhu Xiaoyong. Technology and Applications of Surveying and Mapping for ZY-3 Sattellites[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(10): 1 482-1 491 https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201710031.htm
[11] Reuter H, Nelson A, Jarvis A. An Evaluation of Void-Filling Interpolation Methods for SRTM Data[J]. International Journal of Geographical Information Science, 2007, 21(9): 983-1 008 doi: 10.1080/13658810601169899
[12] Jhee H, Cho H, Kahng H, et al. Multiscale Quadtree Model Fusion with Super-Resolution for Blocky Artefact Removal[J]. Remote Sensing Letters, 2013, 4(4): 325-334 doi: 10.1080/2150704X.2012.729869
[13] Karkee M, Steward B L, Aziz S A. Improving Quality of Public Domain Digital Elevation Models Through Data Fusion[J]. Biosystems Engineering, 2008, 101(3): 293-305 doi: 10.1016/j.biosystemseng.2008.09.010
[14] Yue Linwei, Shen Huanfeng, Yuan Qiangqiang, et al. Fusion of Multi-scale DEMs Using a Regularized Super-Resolution Method[J]. International Journal of Geographical Information Science, 2015, 29(12): 2 095-2 120 doi: 10.1080/13658816.2015.1063639
[15] Moon S, Choi H. Super Resolution Based on Deep Learning Technique for Constructing Digital Elevation Model[C]. AIAA SPACE Forum, Long Beach, California, USA, 2016
[16] Li Xinghua, Shen Huanfeng, Feng Ruitao, et al. DEM Generation from Contours and a Low-Resolution DEM[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 134: 135-147 doi: 10.1016/j.isprsjprs.2017.09.014
[17] Arefi H, Reinartz P. Accuracy Enhancement of ASTER Global Digital Elevation Models Using ICESat Data[J]. Remote Sensing, 2011, 3(7): 1 323-1 343 doi: 10.3390/rs3071323
[18] Yue Tianxiang, Chen Chuanfa, Li Bailian. A High-Accuracy Method for Filling Voids and Its Verification[J]. International Journal of Remote Sensing, 2012, 33(9): 2 815-2 830 doi: 10.1080/01431161.2011.621465
[19] Yue Linwei, Shen Huanfeng, Zhang Liangpei, et al. High-Quality Seamless DEM Generation Blending SRTM-1, ASTER GDEM v2 and ICESat/GLAS Observations[J]. ISPRS Journal of Photogrammetry & Remote Sensing, 2017, 123(21): 20-34 http://www.sciencedirect.com/science/article/pii/S0924271616305135
[20] Pham H, Marshall L, Johnson F, et al. A Method for Combining SRTM DEM and ASTER GDEM2 to Improve Topography Estimation in Regions Without Reference Data[J]. Remote Sensing of Environment, 2018, 210: 229-241 doi: 10.1016/j.rse.2018.03.026
[21] Nuth C, Kaab A. Co-registration and Bias Corrections of Satellite Elevation Data Sets for Quantifying Glacier Thickness Change[J]. Cryosphere, 2011, 5(1): 271-290 doi: 10.5194/tc-5-271-2011
[22] Huber M, Wessel B, Kosmann D, et al. Ensuring Globally the TanDEM-X Height Accuracy: Analysis of the Reference Data Sets ICESat, SRTM and KGPS-Tracks[C]. 2009 Geoscience and Remote Sensing Symposium, Cape Town, South Africa, 2009
[23] Chen Jun, Ban Yifang, Li Songnian. China: Open Access to Earth Land-Cover Map[J]. Nature, 2014, 514(7 523): 434
[24] Hinton G, Salakhutdinov R. Reducing the Dimensionality of Data with Neural Networks[J]. Science, 2006, 313(5 786): 504-507 doi: 10.1126/science.1127647
[25] Li Tongwen, Shen Huanfeng, Yuan Qiangqiang, et al. Estimating Ground-Level PM2.5 by Fusing Satellite and Station Observations: A Geo-Intelligent Deep Learning Approach[J]. Geophysical Research Letters, 2017, 44(11): 985-993 doi: 10.1002/2017GL075710/full
[26] Luedeling E, Siebert S, Buerkert A. Filling the Voids in the SRTM Elevation Model—A TIN-Based Delta Surface Approach[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2007, 62(4): 283-294 doi: 10.1016/j.isprsjprs.2007.05.004
-
期刊类型引用(3)
1. 周永章，陈川，张旗，王功文，肖凡，沈文杰，卞静，王亚，杨威，焦守涛，刘艳鹏，韩枫. 地质大数据分析的若干工具与应用. 大地构造与成矿学. 2020(02): 173-182 . 
百度学术
2. 王叶晨梓，杜震洪，张丰，刘仁义. 面向分片地图的多分辨率格点数据统一存取方法. 浙江大学学报(理学版). 2017(05): 584-590 . 百度学术
3. 朱建章，石强，陈凤娥，史晓丹，董泽民，秦前清. 遥感大数据研究现状与发展趋势. 中国图象图形学报. 2016(11): 1425-1439 . 百度学术
其他类型引用(6)
计量
- 文章访问数: 1376
- HTML全文浏览量: 514
- PDF下载量: 169
- 被引次数: 9
