ANN-Based DEM Fusion and DEM Improvement Frameworks in Regions of Assam and Meghalaya using Remote Sensing Datasets



Abstract Views
522

Downloads
397

Citations

Share

##plugins.themes.bootstrap3.article.sidebar##

Submitted Jun 22, 2022
Published Aug 15, 2022
10.24018/ejgeo.2022.3.4.305

Abstract viewed = 522 times

Table of Contents

##plugins.themes.bootstrap3.article.main##

The Digital Elevation Models (DEMs) are a key and primary input to a large number of modelling processes such as disaster risk monitoring, flood modelling, hydrology, geology, geomorphology, climatology, and environmental study applications. The DEM serves as an important source of topographic information representing the continuous surface of the earth in 3 dimensions with x, y and z coordinates of any point in a grided raster form or a vector TIN form. This study is based on developing a new method using the universal approximation capability of neural networks for the fusion and improvement of L-band and X-band SAR (Synthetic Aperture Radar) DEMs in the complex terrain of Assam and Meghalaya states of the Indian geographic region using the DEM fusion technique. The high-spatial-resolution ALOS PALSAR RTC HR 12.5 m DEM products are used in a fusion framework designed with neural network models. The network adaptively learns the terrain information to produce fused DEM products. The neural network models generate the relationship between the input elevation information from ALOS PALSAR DEMs and precise reference elevation from ICESat-2 spaceborne altimetry as target data. The training and testing data samples are prepared and filtered by checking the correct range of elevation from the toposheets of this region. Different models are used to separately train for the relatively plain valley portion of the Assam region and mountainous portions of the highland Shillong plateau. The obtained fused DEMs from the developed neural net structure are assessed for their quality and accuracy by estimating the RMSE parameter. The fused DEM attains an RMSE value of 7 meters for the complete region which is a significant improvement over the input DEM RMSE of approximately 11 meters. The plain area points and mountainous region points are assessed separately to analyze the predictions from neural nets in the two types of terrains observed in this study site. Moreover, TanDEM-X 90 m DEM is improved using the neural network modeling in the geometric distortions affected areas, which shows an improvement of around 33% overall at the study site. The assessment of plain and mountainous region points for near-ground points shows an improvement of 47% and 55% respectively. The fusion framework designed using the neural network models is an effective and efficient method for obtaining fused DEMs as well as for the improvement of the existing DEMs for complex terrain.

Keywords: Artificial Neural Network (ANN), DEM Fusion, ICESat-2 Altimetry, Interferometry SAR.

References

  1. Miller CL, Laflamme RA. The digital terrain model - Theory & Application. Am Soc Photogramm 1958; XXIV: 11.
     Google Scholar
  2. Maune DF. Digital elevation model technologies and applications The DEM users manual, 2nd edition. Maryland: selbstverl. der American Society for Photogrammetry and Remote Sensing, 2007.
     Google Scholar
  3. Wolf PR, Dewitt BA, Wilkinson BE. Elements of Photogrammetry. 4th ed. McGraw-Hill Education, 2014, 2014. Epub ahead of print 2014. DOI: 10.1128/AAC.03728-14.
     Google Scholar
  4. Fayard F, Méric S, Pottier E. Matching stereoscopic SAR images for radargrammetric applications. In: International Geoscience and Remote Sensing Symposium (IGARSS). 2007, pp. 4364–4367.
     Google Scholar
  5. Jiang C, Douté S, Luo B, et al. Fusion of photogrammetric and photoclinometric information for high-resolution DEMs from Mars in-orbit imagery. ISPRS J Photogramm Remote Sens 2017; 130: 418–430.
     Google Scholar
  6. Yu JH. Interferometric Synthetic Aperture Radar and Radargrammetry for Accurate Digital Elevation Model Generation in New South Wales. PhD. Thesis. The University of New South Wales; 2011.
     Google Scholar
  7. Yuan, F, Zhang, JX, Zhang, L, Gao J. DEM Generation from Airborne LIDAR data. Int Arch Photogramm Remote Sens Spat Inf Sci, 2009; 38: 308-312.
     Google Scholar
  8. Liu X, Zhang Z. Lidar Data Reduction for Efficient and High Quality Dem Generation. XXI Congr Int Soc Photogramm Remote Sens 2008; 3–11.
     Google Scholar
  9. Massonnet D, Feigl KL. Radar interferometry and its application to changes in the earth’s surface. Rev Geophys, 1998; 36: 441–500.
     Google Scholar
  10. Ferretti A, Monti-guarnieri A, Prati C, et al. InSAR Principles: Guidelines for SAR Interferometry Processing and Interpretation. 2007.
     Google Scholar
  11. Fukumori I. Data Assimilation by Models. Satell Altimetry Earth Sci 2001; 237–265.
     Google Scholar
  12. Zhang J. Multi-source remote sensing data fusion: Status and trends. International Journal of Image and Data Fusion 2010; 1: 5–24.
     Google Scholar
  13. Bhardwaj A, Jain K, Chatterjee RS. Generation of high-quality digital elevation models by assimilation of remote sensing-based DEMs. J Appl Remote Sens, 2019; 13: 1.
     Google Scholar
  14. Papasaika H, Kokiopoulou E, Baltsavias E, et al. Fusion of digital elevation models using sparse representations. 2011; 6952 LNCS: 171–184.
     Google Scholar
  15. Bagheri H, Schmitt M, Zhu XX. Fusion of TanDEM-X and Cartosat-1 elevation data supported by neural network-predicted weight maps. ISPRS J Photogramm Remote Sens 2018; 144: 285–297.
     Google Scholar
  16. Kim DE, Liong SY, Gourbesville P, et al. Simple-Yet-Effective SRTM DEM improvement scheme for dense urban cities using ANN and remote sensing data: Application to flood modeling. Water (Switzerland) 2020; 12: 1–14.
     Google Scholar
  17. Papasaika H, Poli D, Baltsavias E. Fusion of digital elevation models from various data sources. Proc Int Conf Adv Geogr Inf Syst Web Serv GEOWS 2009; 117–122.
     Google Scholar
  18. Jawak SD, Luis AJ. Synergetic merging of Cartosat-1 and RAMP to generate improved digital elevation model of Schirmacher Oasis, East Antarctica. Int Arch Photogramm Remote Sens Spat Inf Sci - ISPRS Arch 2014; 40: 517–524.
     Google Scholar
  19. Deo R, Rossi C, Eineder M, et al. Framework for Fusion of Ascending and Descending Pass TanDEM-X Raw DEMs. IEEE J Sel Top Appl Earth Obs Remote Sens, 2015; 8: 3347–3355.
     Google Scholar
  20. Buckley SM, Agram PS, Belz JE, et al. NASADEM: User Guide, https://lpdaac.usgs.gov/documents/592/NASADEM_User_Guide_V1.pdf (2020).
     Google Scholar
  21. Bhardwaj A. Quality Assessment of merged NASADEM products for varied Topographies in India using Ground Control Points from GNSS. MOL2NET 2020, Int Conf Multidiscip Sci 6th Ed 2021; 2–11.
     Google Scholar
  22. Robinson N, Regetz J, Guralnick RP. EarthEnv-DEM90: A nearly-global, void-free, multi-scale smoothed, 90m digital elevation model from fused ASTER and SRTM data. ISPRS J Photogramm Remote Sens, 2014; 87: 57–67.
     Google Scholar
  23. Bhardwaj A. Quality Assessment of Openly Accessible Fused EarthEnvDEM90 DEM and its comparison with MERIT DEM using Ground Control Points for Diverse Topographic Regions. MOL2NET, Int Conf Ser Multidiscip Sci 6th Ed, 2020; 1–8.
     Google Scholar
  24. Yamazaki D, Ikeshima D, Sosa J, et al. MERIT Hydro: A High-Resolution Global Hydrography Map Based on Latest Topography Dataset. Water Resources Research, 2019; 55: 5053–5073.
     Google Scholar
  25. Yamazaki D, Trigg MA, Ikeshima D. Development of a global ~90m water body map using multi-temporal Landsat images. Remote Sens Environ, 2015; 171: 337–351.
     Google Scholar
  26. Elsherbini A, Sarabandi K. Image distortion effects in subsurface SAR imaging of deserts and their correction technique. International Geoscience and Remote Sensing Symposium (IGARSS) 2011: 1075–1078.
     Google Scholar
  27. Woodhouse IH. Introduction to Microwave Remote Sensing. 1st Edition. CRC Press. Epub ahead of print 12 July 2017. DOI: 10.1201/9781315272573.
     Google Scholar
  28. Brown ME, Delgodo Arias S, Neumann T, et al. Applications for ICESat-2 Data: From NASA’s Early Adopter Program. IEEE Geosci Remote Sens Mag, 2016; 4: 24–37.
     Google Scholar
  29. Tian X, Shan J. Comprehensive Evaluation of the ICESat-2 ATL08 Terrain Product. IEEE Trans Geosci Remote Sens, 2021; 59: 8195–8209.
     Google Scholar
  30. Wang C, Zhu X, Nie S, et al. Ground elevation accuracy verification of ICESat-2 data: a case study in Alaska, USA. Opt Express, 2019; 27: 38168.
     Google Scholar
  31. Zhang Y, Pang Y, Cui D, et al. Accuracy Assessment of the ICESat-2/ATL06 Product in the Qilian Mountains Based on CORS and UAV Data. IEEE J Sel Top Appl Earth Obs Remote Sens, 2021; 14: 1558–1571.
     Google Scholar
  32. Bhardwaj A. Investigating the Terrain Complexity from ATL06 ICESat-2 Data for Terrain Elevation and Its Use for Assessment of Openly Accessible InSAR Based DEMs in Parts of Himalaya’s. 2022; 65.
     Google Scholar
  33. Carabajal CC, Harding DJ. ICESat validation of SRTM C-band digital elevation models. Geophys Res Lett, 2005; 32: 1–5.
     Google Scholar
  34. Dandabathula G, Sitiraju SR, Jha CS. Retrieval of building heights from ICESat-2 photon data and evaluation with field measurements. Environ Res Infrastruct Sustain, 2021; 1: 011003.
     Google Scholar
  35. Goud GPS, Bhardwaj A. Estimation of Building Heights and DEM accuracy Assessment using ICESat-2 Estimation of Building Heights and DEM accuracy Assessment using ICESat-2 data products.
     Google Scholar
  36. Hu YH, Hwang JN. Handbook of Neural Network Signal Processing . The Journal of the Acoustical Society of America, 2002; 111(6): 2525–2526. https://doi.org/10.1121/1.1480419.
     Google Scholar
  37. Gurney K. Introduction to neural networks. Taylor and Francis Group; 1997. [Available] DOI: 10.1016/S0140-6736(95)91746-2.
     Google Scholar
  38. Anderson JA. Introduction to neural networks. 8th ed. 1994. [Available] DOI: 10.5860/choice.35-5700.
     Google Scholar
  39. Kanungo DP, Arora MK, Sarkar S, et al. A comparative study of conventional, ANN black box, fuzzy and combined neural and fuzzy weighting procedures for landslide susceptibility zonation in Darjeeling Himalayas. Eng Geol, 2006; 85: 347–366.
     Google Scholar
  40. Kavzoglu T, Mather PM. The use of backpropagating artificial neural networks in land cover classification. Int J Remote Sens, 2003; 24: 4907–4938.
     Google Scholar
  41. Demuth H, Beale M. Neural Network Toolbox Version4. Networks, 2002; 24: 1–8.
     Google Scholar
  42. Kulp SA, Strauss BH. CoastalDEM: A global coastal digital elevation model improved from SRTM using a neural network. Remote Sens Environ, 2018; 206: 231–239.
     Google Scholar
  43. Kampüs K. Estimation of Unknown Height With Artificial Neural Network on Digital Terrain Model. IsprsOrg, 2002; 2: 115–118.
     Google Scholar
  44. Nguyen NS, Kim DE, Jia Y, et al. Application of Multi-Channel Convolutional Neural Network to Improve DEM Data in Urban Cities. Technologies, 2022; 10: 61.
     Google Scholar
  45. Girohi P, Bhardwaj A. A Neural Network based Fusion Approach for Improvement of SAR Interferometry based Digital Elevation Models in Plain and hilly terrain of India. To be published in AI MDPI. Preprint 2022. [Available] DOI: 10.20944/preprints202206.0013.v1.
     Google Scholar
  46. Wessel B. TanDEM-X ground segment DEM products specification document. Public Doc TD-GS-PS-0021 2018; TD-GS-PS-0: 49.
     Google Scholar
  47. Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim Res, 2005; 30: 79–82.
     Google Scholar
  48. Kirthiga SM, Patel NR. Impact of updating land surface data on micrometeorological weather simulations from the WRF model. Atmosfera, 2018; 31: 165–183.
     Google Scholar