Inhalt

•  
   Abstrakt
•  
   Abstract
•  
   Acknowledgements
•  
   Preface
•  
   Contents
•  
   List of Figures
•  
   List of Tables
•  
  1 Introduction
•  
   1.1 Motivation
•  
  1.2 Problem Description
•  
   1.2.1 Terrain Classification
•  
   1.2.2 Vegetation Detection
•  
   1.3 Goald of this Thesis
•  
   1.4 Novel Contributions of the Thesis
•  
   1.4.1 Fitting plane Algorithm-based Depth Correction for Tyzx DeepSea Stereoscopic Imaging
•  
   1.4.2 Vegetation Indices Applied for Vegetation Detection
•  
   1.4.3 2D/3D Feature Fusion for Vegetation Detection
•  
   1.4.4 General Vegetation Detection Using an Integrated Vision System
•  
   1.4.5 Spreading Algorithm for Efficient Vegetation Detection
•  
   1.4.6 A Novel Approach for a Double-Check of Passable Vegetation Detection in Autonomous Ground Vehicles
•  
   1.4.7 Terrain Classification Based on Structure for Autonomous Navigation in Complex Environments
•  
   1.4.8 A Novel Approach of Terrain Classification for Outdoor Automobile Navigation
•  
   1.5 Document Structure
•  
   1.6 Publications
•  
   2 Fundamentals
•  
   2.1 The Experimental Platform AMOR
•  
  2.2 Light Detection And Ranging (LiDAR)
•  
   2.2.1 Optical Triangulation for 3D Digitizing
•  
   2.2.2 Laser Pulse Time-of-flight
•  
   2.2.3 Laser Phase-Shift Range Finder
•  
   2.2.4 Laser Scanner SICK LMS221
•  
   2.3 Structured Light
•  
   2.4 The MultiCam
•  
   2.5 Stereoscopic Imaging
•  
   2.5.1 Fitting Plane Algorithm-based Depth Correction for Tyzx DeepSea Stereoscopic Imaging
•  
   2.5.1.1 Introduction
•  
   2.5.1.2 Planar Surface for Scene Understanding
•  
   2.5.1.3 Fitting Plane Algorithm
•  
   2.5.1.4 Experiments and Results
•  
   2.5.1.5 Conclusion
•  
   2.6 Multi-spectral Imaging
•  
   3 Vegetation Indices Applied for Vegetation Detection
•  
   3.1 Related Work
•  
   3.1.1 Ratio Vegetation Index
•  
   3.1.2 Normalized Difference Vegetation Index
•  
   3.1.3 Perpendicular Vegetation Index
•  
   3.1.4 Difference Vegetation Index
•  
   3.1.5 Soil-Adjusted Vegetation Index
•  
   3.1.6 Modified Soil Adjusted Vegetation Index
•  
   3.2 A Novel Vegetation Index : Modification of Normalized Difference Vegetation Index
•  
   3.2.1 Derivation of Novel Index
•  
   3.3 Experiments and Results
•  
   3.4 Conclusion
•  
   4 2D-3D Feature Fusion-based Vegetation Detection
•  
   4.1 Related Work
•  
   4.2 2D/3D Mapping
•  
   4.3 3D point cloud analysis
•  
   4.3.1 Scatter Feature Extraction
•  
   4.4 Colour Descriptors
•  
   4.5 Support Vector Machine
•  
   4.6 Experiments and Results
•  
   4.7 Conclusion
•  
   5 General Vegetation Detection Using an Integrated Vision System
•  
   5.1 System Set-Up
•  
   5.2 Spatial Features
•  
   5.3 Vegetation Index Calculation
•  
   5.4 Colour and Texture Descriptors
•  
   5.5 Experiments and Results
•  
   5.6 Conclusion
•  
   6 Spreading Algorithm for Efficient Vegetation Detection
•  
   6.1 Introduction
•  
   6.2 Discussion on Vegetation Indices
•  
   6.3 Visual Features for Scene Understanding
•  
   6.3.1 Opponent Color Space
•  
   6.3.2 Unstructured Texture
•  
   6.4 Spreading Algorithm
•  
   6.5 Experiments and Results
•  
   6.6 Conclusion
•  
  7 A Novel Approach for a Double-Check of Passable Vegetation Detection in Autonomous Ground Vehicles
•  
   7.1 Introduction
•  
  7.2 Multi-spectral-based Vegetation Detection
•  
   7.2.1 Standard Form of Vegetation Index
•  
   7.2.2 Modification Form of Vegetation Index
•  
   7.2.3 Convex Combination of Vegetation Indices
•  
   7.3 System Design
•  
   7.4 A Double-Check for Passable Vegetation Detection
•  
   7.5 Experiments and Results
•  
   7.6 Conclusions
•  
   8 Terrain Classification Based on Structure for Autonomous Navigation in Complex Environments
•  
   8.1 Introduction
•  
   8.2 Methodology
•  
   8.2.1 Efficient Graph-based Segmentation Technique
•  
  8.2.2 Feature Extraction
•  
   8.2.2.1 Neighbour Distance Variation Inside Edgeless Regions
•  
   8.2.2.2 Conditional Local Point Statistics
•  
   8.2.3 Support Vector Machine
•  
   8.3 Experiments and Results
•  
   8.4 Conclusion
•  
  9 A Novel Approach of Terrain Classification for Outdoor Automobile Navigation
•  
   9.1 Introduction
•  
   9.2 Related Works
•  
   9.3 2D/3D Coarse Calibration
•  
  9.4 Feature-based Classification
•  
   9.4.1 Depth Image Segmentation
•  
  9.4.2 2D/3D Feature Fusion
•  
   9.4.2.1 3D Features
•  
   9.4.2.2 2D Features
•  
   9.5 Experiments and Results
•  
   9.6 Conclusion
•  
  10 Conclusions
•  
   10.1 Summary
•  
   10.2 Discussion
•  
   10.3 Direction for Future Work
•  
   Appendix A - Expert Concerns and Rebuttal
•  
   References