Knowledge based building facade reconstruction from laser point clouds and images : dissertation
Material type: TextSeries: ITC dissertation ; No.168Publication details: Netherlands International Inst. for Geo-information science and Earth Observation (ITC) 2010Description: ix,119pISBN:- 9061642884
- 711.4 PU
Item type | Current library | Collection | Call number | Status | Notes | Date due | Barcode | Item holds | |
---|---|---|---|---|---|---|---|---|---|
Book | CEPT Library | Faculty of Planning | 711.4 PU | Available | Status:Catalogued;Bill No:GRATIS | 006514 |
CONTENTS 1 Introduction 1 1.1 State-of-the-art of terrestrial laser scanning 3 1.2 Related works6 1.2.1 Overview 6 1.2.2 Fruch et al. 2005 8 1.2.3 Cornells ct al. 2008 8 1.2,4 Ripperda 20089 1.2.5 Backer 200910 1.3 Method overview 11 1.4 Structure of the thesis 12 2 Knowledge engineering and reasoning15 2.1 Knowledge engineering16 2.1.1 Assembling the knowledge16 2.1.2 Decide on a vocabulary19 2.1.3 Encode general knowledge 20 2.1.4 The hierarchical composition 22 2.2 Reasoning with the knowledge 22 2.2.1 The proof tree22 2.3 Managing uncertainty 26 2.3.1 Describing the uncertainty 27 2.3.2 Making expected decisions 29 2.4 Concluding remarks 31 3 Feature extraction 33 3.1 Preprocessing 33 3.1.1 Spatial indexing 33 3.1.2 Extracting points of interest 34 3.2 Extraction of geometric features36 3.2.1 Flat surfaces 36 3.2.2 Curved surfaces 38 3.3 Kxtraction of semantic features39 3.3.1 Solid features extraction 40 3.3.2 Hole-based window extraction 42 3.4 Discussion 43 4 Geometric reconstruction 45 4.1 Polygon fitting45 4. 1.1 Least squares fitting45 4.1.2 Convex polygon and concave polygon fitting 46 4.1.3 Minimum bounding rectangle fitting 48 4.2 B-spline surface fitting 48 4.2.1 The B-spline curve and surface 49 4.2.2 B-spline surface approximation 50 4.3 Hypotheses for parts without laser data51 4.4 Results and Discussion 51 4.4.1 Flat surfaces51 4.4.2 Curved surfaces 53 5 Model refinement with imagery 57 5.1 Method overview 58 5.2 Registration 58 5.2.1 Perspective Conversion 59 5.2.2 Spatial Resection 62 5.2.3 Relative Orientation 62 5.3 The model refinement64 5.3.1 Extraction of Significant Lines from Jmages 65 5.3-2 Matching Model Edges with linage Lines 65 5.3.3 Refinement Strategy 67 5.4 Test cases 67 5-4.1 The restaurant house 67 5.4.2 The town hall 68 5.4.3 The wall with high windows 69 5.4.4 Summary71 5.5 Conclusions and outlook 72 6 Texture mapping 73 6.1 Selecting texture images 74 6.1.1 Optimal image selection 75 6.1.2 Occlusion removal75 6.2 Calculating texture coordinates 77 6.3 Results and discussion 78 6.3.1 The three joined 78 6.3.2 The house with a balcony 79 6.3.3 The curved walls 80 6.3.4 Discussion 80 7 Method evaluation 83 7.1 The reconstruction approaches 83 7.1.1 Our approach 84 7.1.2 The traditional approach 85 7.2 The Vlaardingen case 86 7.3 The Enschede case 94 7.4 The Esslingen case 95 7.5 Conclusions 96 8 Conclusions and recommendations 103 8.1 Conclusions103 8.2 Recommendations104 Bibliography 106 List of publications 111 List of Figures 113 List of Tables 115
There are no comments on this title.