一個LiDAR DTM的可靠度指標及其初步測試

A RELIABILITY INDEX of LiDAR DTM AND ITS PRELIMINARY TESTS

10.6652/JoCICHE.201910_31(6).0007

蔡展榮(Jaan-Rong Tsay)；吳品融(Ping-Rong Wu)

光達 ； 數值地形模型 ； 可靠度 ； LiDAR ； digital terrain model (DTM) ； reliability

中國土木水利工程學刊

31卷6期（2019 / 10 / 01）

589 - 603

繁體中文

本文設計並提出一個LiDAR DTM 可靠度的簡易指標ξ，計算每一個LiDAR DTM 網 格點高程的ξ值（0 ≤ξ≤ 1）。所謂「可靠度」就是資料的可靠程度，越高的ξ值代表這一個LiDAR DTM網格點的高程值越可靠。ξ= 0表示這一個LiDAR DTM 網格點的高程是不可靠的，因為在其地表起伏相關區域裡無光達地面點。在未來完成完整的研究並確認可行性之後，建議內政部提供LiDAR DTM的資料內容宜包括每一個網格點的三維地面坐標（X, Y, Z）及其合適的可靠度指標值共4筆數據，俾供使用者判斷每一個LiDAR DTM網格點高程值的可靠程度。本文初步的實驗測試成果顯示：在相同的網格點ξ值之下，坡度等級越大，網格點高程較差dZ值的範圍也越大，例如在60%≤ξ≤80%下，較平坦的坡度等級1及2的高程較差dZ落在[-10.44 m, 12.68 m]範圍，最陡峭的坡度等級4則是-19.12 m≤ dZ ≤ 12.28 m。因被國家規定管制，高解析度LiDAR DTM（及其原始LiDAR）資料無法取得，又實際地真資料取得困難且費時，故本篇研究成果有限，未達原本期望。儘管如此，還是希望藉本文拋磚引玉，讓大家重視LiDAR DTM可靠度，以保障大眾生命財產之安全，也誠盼中央主管機關內政部能重視並啟動這項研究及應用。

This paper designs and proposes a simple reliability index ξ of LiDAR DTM. Aξ-value (0 ≤ξ≤ 1) is computed for each grid point of LiDAR DTM. The so-called ＂reliability＂ is an index to express to what extent the data is reliable. A higherξ-value denotes that the grid point has a more reliable elevation. ξ= 0 indicates that the elevation of the grid point is unreliable because there is not any LiDAR terrain point inside the topography correlation area. After further study is completed in the coming years and the applicability of reliability index for LiDAR DTM is confirmed, the Ministry of Interior (MOI) is strongly suggested to supply the 3D ground coordinates (X, Y, Z) and a proper reliability index value of each grid point in a LiDAR DTM to users so that users can know which grid points have reliable elevations. Our preliminary test results demonstrate that the elevations of grid points in steeper areas will have larger errors, if they have the sameξ-values. In the case with 0.6 ≤ξ≤ 0.8, the elevation difference dZ between check data and computed data by LiDAR terrain points ranges from -10.44 m to 12.68m for those grid points in the areas with the slope level 1 and 2. For those grid points in the steepest areas with the slope level 4, their dZ-values range from -19.12 m to 12.28 m. High resolution, namely 1mx1m, LiDAR DTM and its original data of LiDAR terrain points are not available due to our national regulations. Moreover, acquisition of ground truth data in study areas is difficult and time-consuming. Therefore, the study results fail to meet our expectation. Nevertheless, it would like to inspire each and every one of you to consider the reliability of LiDAR DTM an important index. Also, the Ministry of Interior is strongly suggested to start the related researches and developments in order to be able to then safeguard the safety of people's lives and property.

1. IKONOS DSM/DTM, MAPMART, http://www.mapmart.com/Default.aspx.
2. DigitalGlobe, http://www.digitalglobe.com/.
3. Geoeye DSM/DTM, MAPMART, http://www.mapmart.com/Default.aspx.
4. GEO/TU Wien and Trimble Germany, SCOP++ Manual for Version 5.5 and higher. https://photo.geo.tuwien.ac.at/ media/filer_public/c1/24/c1248934-763c-4f2a-9afa-fc122dc3b953/referencemanual_scop_553.pdf (2014).
5. ALOS DSM, ALOS Pegasus, http://pegasus.alos-pasco.com/.
6. Geo Elevation 30, Astrium, http://www.astrium-geo.com/.
7. 國土測繪法 (2007)。
8. (2010).Inpho GmbH, Institute of Photogrammetry and Remote Sensing (Institut fuer Photogrammetrie und Fernerkundung, IPF) of the TU Vienna, SCOP++ Manual, Stuttgart / Vienna, pp. 367 (2010)..
9. ASTRIUM, http://www.astrium-geo.com/.
10. 數值地形模型成果資料流通供應要點 (2013)。
11. NEXTMap World 30 meter DSM, Intermap, http://www.mapmart.com/Default.aspx.
12. Worldview Elevation Suite, DigitalGlobe, http://www.digitalglobe.com/.
13. Geo Elevation 10, Astrium, http://www.astrium-geo.com/.
14. ASTER GDEM, METI and NASA, http://gdem.ersdac.jspacesystems.or.jp/index.jsp.
15. Baarda W.(1968).A testing procedure for use in geodetic networks.Delft, The Netherlands:Netherlands Geodetic Commission.
16. Flood, M., ASPRS Guidelines Vertical Accuracy Reporting for Lidar Data, ASPRS Lidar Committee (PAD), http://www.asprs.org/a/society/committees/standards/Vertical_Accuracy_Reporting_for_Lidar_Data.pdf (2004).
17. Guo, Q.,Li, W.,Yu, H.,Alvarez, O.(2010).Effects of topographic variability and lidar sampling density on several DEM interpolation methods.Photogrammetric Engineering & Remote Sensing,76(6),701-712.
18. Heurich, M., Mette, T., Krzystek, P., Hajnsek, I., Aulinger, T. and Papathanassiou, K. P., Comparison of DTM, DSM and DCM derived from LIDAR, InSAR and Digital Camera over a mixed mountain forest in central Europe, http://cears.fw.vt.edu/silviscan/presentations/Author60.pdf (2013).
19. Höhle, J. and Potuckova, M., Assessment of the Quality of Digital Terrain Models, European spatial data research, EurSDR. http://www.eurosdr.net/publications/60.pdf (2011).
20. Kraus, K.(2007).Photogrammetry: Geometry from Images and Laser Scans.Walter de Gruyter.
21. Kraus, K.,Mikhail, E.M.(1972).Linear Least-Squares Interpolation, Twelfth Congress of the International Society of Photogrammetry, Ottawa, Canada, July 23 – August 5, 1972.Presented paper, Commission III
22. Moritz, H.,Suenkel, H.(1978).Approximation Methods in Geodesy.Karlsruhe, Germany:Herbert Wichmann Verlag.
23. Papasaika, H.,Kokiopoulou, E.,Baltsavias, E.,Schindler, K.,Kressner, D.(2011).Fusion of Digital Elevation Models Using Sparse Representations.Proceedings of the Photogrammetric Image Analysis,Munich, Germany:
24. Pfeifer, N.(2008).Pfeifer, N., I.P.F. team, DSM/DTM Filtering. International School on Lidar Technology 2008, IIT Kanpur, India (2008)..
25. Wolf, P.,DeWitt, B. A.(2000).Elements of Photogrammetry with Applications in GIS.McGraw-Hill.
26. Бронштейн, И. Н.,Семендяев, К. А.(1986).數學手冊.蘇聯科學出版社.
27. 內政部(2014)。內政部，「103 年度數值地形模型相關計畫成果維護管理及整合服務工作案」期末報告 (2014)。
28. 內政部(2013)。內政部，「102 年度數值地形模型相關計畫成果維護管理及整合服務工作案」期末報告 (2013)。
29. 內政部，一千分之一數值航測地形圖測製作業規定，(https://bmap.nlsc.gov.tw/files/topographic/OperatingRequire.pdf)，附件 3，內政部：台北 (2011a)。
30. 內政部，建置都會區一千分之一數值航測地形圖作業工作手冊，附錄 3「一千分之一數值航測地形圖測製作業規定」的附件 3，https://www.nlsc.gov.tw/Law/Standard?level=108，內政部：台北 (2011b)。
31. 吳品融(2018)。國立成功大學測量及空間資訊學系。
32. 吳品融,蔡展榮,葉峻輔(2018)。表達空載光達 DTM 可靠度的一個新指標 ξ。第 37 屆測量及空間資訊研討會，會場 2 (R2-116) 「空間資訊與行動測繪 (二) 系統設計規劃」
33. 陳品清(1982)。國立成功大學航空測量研究所。
34. 蔡展榮(1998)。取樣定理內的正交級數之訊號表達精度與收斂性。測量工程，40(3)，45-70。
35. 蔡展榮(1986)。國立成功大學航空測量研究所。