| Acknowledgement | 第1-14页 |
| 摘要 | 第14-20页 |
| Abstract | 第20-23页 |
| Contents | 第23-26页 |
| List of Figures | 第26-32页 |
| List of Tables | 第32-33页 |
| Chapter(1) Introduction | 第33-44页 |
| ·Purpose and significance of this study | 第33-38页 |
| ·Summary and conclusions | 第38-39页 |
| ·Structure of the thesis | 第39-44页 |
| Chapter(2) Literature review | 第44-59页 |
| ·Related work | 第44-54页 |
| ·Region growing method | 第45-46页 |
| ·Clustering of features based method | 第46-48页 |
| ·The model fitting method(RANSAC algorithm) | 第48-51页 |
| ·Evaluate the spurious results from RANSAC algorithm | 第50-51页 |
| ·A comparison between the three distinct methods | 第51-54页 |
| ·Summary and conclusions | 第54-59页 |
| Chapter(3) Research framework | 第59-67页 |
| ·Introduction | 第59-60页 |
| ·Research objectives and motivations | 第60-61页 |
| ·Research problems and questions | 第61-67页 |
| ·Data acquisitions | 第61-62页 |
| ·Preparing process | 第62-63页 |
| ·Solving the main problem | 第63-66页 |
| ·Quality measurement | 第66-67页 |
| Chapter (4) An improved segmentation approach for planar surfaces called "SEQ-NV-RANSAC" | 第67-76页 |
| ·General workflow | 第67-68页 |
| ·Neighbours groups and fitting | 第68-71页 |
| ·Segmentation based on cluster features | 第71-72页 |
| ·SEQ-NV-RANSAC approach | 第72-76页 |
| Chapter (5) Field works and analysis the experimental results | 第76-97页 |
| ·Data description | 第76-80页 |
| ·Analysis acquisition data | 第80-82页 |
| ·Registration using targets-based registration tool | 第82-85页 |
| ·Partitioning the raw data based on the requirements (objectives) | 第85-95页 |
| ·Delete the clearly non-beneficial point clouds | 第85-86页 |
| ·Stairs data test:as ideal example for parallel-gradual planar surfaces | 第86-89页 |
| ·Column data test:as example of different orientation for planar surfaces | 第89-91页 |
| ·Full facade data test:as example for complex facade with massive 3D point clouds | 第91-95页 |
| ·Summary and conclusions | 第95-97页 |
| Chapter (6) An extension of "SEQ-NV-RANSAC" approach to avoid bad-segmentation cases | 第97-121页 |
| ·Modified workflow | 第97-98页 |
| ·Under-segmentation problem | 第98-104页 |
| ·Over-segmentation problem | 第104-121页 |
| ·Cluster groups based on the normal vector directions(NV) | 第106-108页 |
| ·Cluster groups based on average summation of perpendicular(AveSumDis) | 第108页 |
| ·Re-Merging groups based on(MinPointNo)into intersection zone | 第108-113页 |
| ·Experimental results and analysis | 第113-121页 |
| Chapter(7) Interprets linear features(edges)for planar surfaces | 第121-152页 |
| ·Introduction | 第121-123页 |
| ·Return-back the rejection points | 第123-129页 |
| ·Points of edges | 第129-144页 |
| ·Intersection edges(IntEdges)algorithm | 第129-135页 |
| ·Free boundary points(FreeBoundaryPoints)algorithm | 第135-140页 |
| ·Merge the edges points(MergeEdPo) | 第140-144页 |
| ·Extracting all edges from existing planar surfaces using"Seq-RANSAC-Edges"approach | 第144-152页 |
| ·Experimental results and analysis | 第148-152页 |
| Chapter(8) Conclusions and future work | 第152-158页 |
| ·Introduction | 第152-153页 |
| ·Conclusions | 第153-157页 |
| ·Trends for future work | 第157-158页 |
| References | 第158-173页 |
| Publications | 第173-174页 |
| Index(Ⅰ)-The color figures | 第174-197页 |
