簡易檢索 / 詳目顯示

回結果列表
研究生： 蘇柏瑋
Su, Bo-Wei
論文名稱： 以高解析衛星影像產製之數值地表模型與真實正射影像進行LOD-1房屋模型重建
LOD-1 Building Model Reconstruction from HRSI Derived DSM and True-orthoimage
指導教授： 饒見有
Rau, Jiann-Yeou
學位類別： 碩士
Master
系所名稱： 工學院 - 測量及空間資訊學系
Department of Geomatics
論文出版年： 2017
畢業學年度： 105
語文別： 英文
論文頁數： 82
中文關鍵詞： LOD-1房屋模型 、物件導向式影像分析 、高解析衛星影像 、數值地表模型 、真實正射影像
外文關鍵詞： LOD-1 Building Models, OBIA, HRSI, DSM, True-orthoimage
相關次數： 點閱：168  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 三維房屋模型在都市的環境研究裡扮演著重要的角色，相關的應用包含都市計劃、災害管理、災損評估、危險評估、災害模擬等。本研究的目標為開發出一套低成本的流程，以高解析衛星影像產製之數值地表模型與真實正射影像來重建CityGML LOD-1房屋模型。本研究困難的地方在於建物通常擁有各式各樣的顏色、紋理與形狀等。許多的問題讓萃取房屋區域的議題充滿挑戰性。本研究整合OSM的道路向量資料、數值地表模型與真實正射影像來進行物件導向式影像分析，以萃取出建物區域。其中，數值地表模型與真實正射影像皆從Pléiades高解析衛星影像產製而得。本研究主要的處理單位為由多解析影像切割方法所產製的影像物件，並推導出每個影像物件的物件屬性，例如：光譜、幾何屬性等，以進行規則式分類。在萃取完建物區域之後，由於原始的建物邊界為不規則狀，因此對每個房屋邊界進行規則化。最後，將每個建物區域賦予對應的平均高度，而能產製出LOD-1房屋模型。除此之外，本研究採用三組不同收斂角的衛星立體像對產製出三組資料，以測試收斂角是否對成果產生影響。精度評估的結果顯示出當收斂角愈大，則成果的精度隨之減少。在本研究裡，萃取建物區域最佳的整體精度與kappa值分別為87.33%與0.73。高程精度為2.83公尺，符合CityGML LOD-1的高度要求。

    Three-dimensional building model is essential for city environment studies, such as urban planning, disaster management, loss estimation, risk modelling and assessment, disaster simulation, etc. The goal of this study is to develop a low cost workflow for the reconstruction of CityGML LOD-1 (level-of-detail # 1) building models from Digital Surface Model (DSM) and true-orthoimage derived from High-Resolution Satellite Imagery (HRSI). The difficulties for this task is majorly due to buildings possessing various colors, textures and shapes of boundary. Several factors still pose challenges that interfere with building footprint extraction. In the research, Object-Based Image Analysis (OBIA) is conducted to extract building footprints by integrating road vector from OpenStreetMap (OSM), DSM and true-orthoimage. In which, the DSM and true-orthoimage are both derived from HRSI of Pléiades. The major processing unit is image object generated by multiresolution image segmentation. Several object features such as spectral value and geometry can be derived based on the image objects, which are used to develop rule sets for classification. After building footprint extraction, building footprints are regularized due to the irregular building boundaries. Finally, LOD-1 building models can be extruded by assigning the mean height to the corresponding building footprints. In addition, three data sets generated from stereo-pair with different convergent angles are tested to know whether convergent angle has influence on the result or not. Accuracy assessment shows that as convergent angle increases, the accuracy of the result decreases. In the research, the best overall accuracy and kappa value of building footprint extraction are 87.33% and 0.73, respectively. The height accuracy is 2.83 meters, which meets the LOD-1 height requirement of CityGML.

    摘要 I Abstract II 致謝 III CATALOG IV List of Tables VII List of Figures VIII Chapter 1 Introduction 1 1.1 Background 1 1.2 Motivation 3 1.3 Literature Review 5 1.3.1 3D Building Model Reconstruction 5 1.3.2 Image Classification 6 1.4 Comparisons with Current Approaches 7 1.5 Paper Structure 8 Chapter 2 Study Area and Material 10 2.1 Study Area 10 2.2 Material 10 2.2.1 True-orthoimage 13 2.2.2 DSM 14 2.2.3 OpenStreetMap 15 Chapter 3 Methodology 17 3.1 Workflow 17 3.2 Data Pre-processing 19 3.2.1 Generation of DSM and True-orthoimage 19 3.2.2 Kuwahara Filter 20 3.2.3 Road Vector from OSM 21 3.3 Image Segmentation 23 3.4 Multiresolution Segmentation 24 3.4.1 Procedure 24 3.4.2 Scale Parameter 25 3.4.3 Shape and Compactness Parameters 25 3.5 Object Features 26 3.5.1 Vegetation Index 27 3.5.2 Object Height Model 28 3.5.3 Brightness 30 3.5.4 Relative border 30 3.5.5 Area 30 3.6 Determination of Threshold 31 3.7 Classification Methods 32 3.8 Building Footprint Extraction 34 3.8.1 Classification of Vegetation 34 3.8.2 Classification of Ground 35 3.8.3 Refinement of Building Footprint 36 3.8.4 Segmentation for Individual Building Footprint 38 3.8.5 Building Footprint Regularization 39 3.9 LOD-1 building model reconstruction 40 Chapter 4 Case studies and Analysis 41 4.1 Data Pre-processing 41 4.1.1 Kuwahara Filter 41 4.1.2 Removal of Road Area 42 4.2 Multiresolution Segmentation 43 4.3 Classification 45 4.3.1 Vegetation 45 4.3.2 Ground 47 4.4 Building Footprint refinement 51 4.5 Comparison of road vector from OSM 56 4.6 Comparison between different data sets 58 4.6.1 Area comparison 58 4.6.2 Comparison of Height correctness in DSMs 62 4.7 Height Accuracy Assessment 67 4.8 Assessment of Building Footprint Regularization 68 4.9 LOD-1 Building Model 70 Chapter 5 Conclusions and Suggestions 74 5.1 Building Footprint Extraction 74 5.2 Convergent Angle of Stereo-pair 75 5.3 Building Footprint Regularization 75 5.4 LOD-1 Building Model 75 Chapter 6 Future works 77 6.1 Tri-stereo imagery 77 6.2 OSM data 77 6.3 Indices 78 6.4 Regularization 78 6.5 Accuracy assessment 78 References 79

    Albert, J., Bachmann, M., Hellmeier, A., 2003. Zielgruppen und anwendungen für digitale stadtmodelle und digitale geländemodelle. Erhebungen im Rahmen der SIG 3D der GDI NRW.
    ASTRIUM, A.E., 2012. Company (2012). PLEIADES imagery user guide.
    Baatz, M., Schäpe, A., 2000. Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation. Angewandte geographische informationsverarbeitung XII 58, 12-23.
    Baillard, C., Zisserman, A., 2000. A plane-sweep strategy for the 3D reconstruction of buildings from multiple images. International Archives of Photogrammetry and Remote Sensing 33, 56-62.
    Biljecki, F., Ledoux, H., Stoter, J., 2017. Generating 3D city models without elevation data. Computers, Environment and Urban Systems 64, 1-18.
    Blaschke, T., 2010. Object based image analysis for remote sensing. ISPRS journal of photogrammetry and remote sensing 65, 2-16.
    Blaschke, T., Lang, S., Lorup, E., Strobl, J., Zeil, P., 2000. Object-oriented image processing in an integrated GIS/remote sensing environment and perspectives for environmental applications. Environmental information for planning, politics and the public 2, 555-570.
    Brenner, C., 2000. Towards fully automatic generation of city models. International Archives of Photogrammetry and Remote Sensing 33, 84-92.
    Brenner, C., 2005. Building reconstruction from images and laser scanning. International Journal of Applied Earth Observation and Geoinformation 6, 187-198.
    Drǎguţ, L., Tiede, D., Levick, S.R., 2010. ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data. International Journal of Geographical Information Science 24, 859-871.
    eCognition, 2014. eCognition Developer Reference Book. München, Germany: Trimble Germany GmbH.
    Elberink, S.O., Vosselman, G., 2009. Building reconstruction by target based graph matching on incomplete laser data: Analysis and limitations. Sensors 9, 6101-6118.
    Gröger, G., Plümer, L., 2012. CityGML–Interoperable semantic 3D city models. ISPRS Journal of Photogrammetry and Remote Sensing 71, 12-33.
    Gröger, G., Kolbe, T., Nagel, C., Häfele, K., 2012. OGC city geography markup language (CityGML) encoding standard, Version 2.0, OGC doc no. 12-019. Open Geospatial Consortium.
    Gribov, A., 2015. Searching for a compressed polyline with a minimum number of vertices. arXiv preprint arXiv:1504.06584.
    Gruen, A., Steidler, F., Wang, X., 2003. Generation and visualization of 3D-city and facility models using CyberCity Modeler (CC-Modeler™). Paper der CORP, Wien.
    Gruen, A., Wang, X., 1998. CC-Modeler: a topology generator for 3-D city models. ISPRS Journal of Photogrammetry and Remote Sensing 53, 286-295.
    Haala, N., Brenner, C., 1998. Interpretation of urban surface models using 2D building information. Computer vision and image understanding 72, 204-214.
    Haala, N., Brenner, C., 1999. Virtual city models from laser altimeter and 2D map data. Photogrammetric Engineering and Remote Sensing 65, 787-795.
    Huang, X., Wen, D., Li, J., Qin, R., 2017. Multi-level monitoring of subtle urban changes for the megacities of China using high-resolution multi-view satellite imagery. Remote Sensing of Environment 196, 56-75.
    Huete, A., Didan, K., Miura, T., Rodriguez, E.P., Gao, X., Ferreira, L.G., 2002. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote sensing of environment 83, 195-213.
    Jabari, S., Zhang, Y., 2013. Very high resolution satellite image classification using fuzzy rule-based systems. Algorithms 6, 762-781.
    Johnson, B., Xie, Z., 2011. Unsupervised image segmentation evaluation and refinement using a multi-scale approach. ISPRS Journal of Photogrammetry and Remote Sensing 66, 473-483.
    Ledoux, H., Meijers, M., 2011. Topologically consistent 3D city models obtained by extrusion. International Journal of Geographical Information Science 25, 557-574.
    Merrell, P., Akbarzadeh, A., Wang, L., Mordohai, P., Frahm, J.-M., Yang, R., Nistér, D., Pollefeys, M., 2007. Real-time visibility-based fusion of depth maps, Computer Vision, 2007. ICCV 2007. IEEE 11th International Conference on. IEEE, pp. 1-8.
    Meyer, G.E., Neto, J.C., 2008. Verification of color vegetation indices for automated crop imaging applications. Computers and electronics in agriculture 63, 282-293.
    Motohka, T., Nasahara, K.N., Oguma, H., Tsuchida, S., 2010. Applicability of green-red vegetation index for remote sensing of vegetation phenology. Remote Sensing 2, 2369-2387.
    Papari, G., Petkov, N., Campisi, P., 2007. Artistic edge and corner enhancing smoothing. IEEE Transactions on Image Processing 16, 2449-2462.
    Qin, R., 2014. Change detection on LOD 2 building models with very high resolution spaceborne stereo imagery. ISPRS Journal of Photogrammetry and Remote Sensing 96, 179-192.
    Qin, R., 2016. RPC stereo processor (RSP)–a software package for digital surface model and orthophoto generation from satellite stereo imagery. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 77-82.
    Rothermel, M., Wenzel, K., Fritsch, D., Haala, N., 2012. Sure: Photogrammetric surface reconstruction from imagery, Proceedings LC3D Workshop, Berlin.
    Suveg, I., Vosselman, G., 2004. Reconstruction of 3D building models from aerial images and maps. ISPRS Journal of Photogrammetry and remote sensing 58, 202-224.
    Tsai, V.J., 2006. A comparative study on shadow compensation of color aerial images in invariant color models. IEEE transactions on geoscience and remote sensing 44, 1661-1671.
    Vu, T.T., Yamazaki, F., Matsuoka, M., 2009. Multi-scale solution for building extraction from LiDAR and image data. International Journal of Applied Earth Observation and Geoinformation 11, 281-289.
    Whiteside, T., Ahmad, W., 2005. A comparison of object-oriented and pixel-based classification methods for mapping land cover in northern Australia, Proceedings of SSC2005 Spatial intelligence, innovation and praxis: The national biennial Conference of the Spatial Sciences Institute, pp. 1225-1231.

    無法下載圖示 校內：2022-09-10公開
    校外：不公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE