| 作者简历 | 第8-11页 |
| 摘要 | 第11-14页 |
| Abstract | 第14-17页 |
| Chapter 1 General introduction | 第21-36页 |
| 1.1 Definitions of some crucial terms | 第21-23页 |
| 1.1.1 Typical geo-environmental information | 第21页 |
| 1.1.2 New remote sensing data | 第21-23页 |
| 1.1.3 Intelligent identification | 第23页 |
| 1.2 Background and motivation | 第23-24页 |
| 1.3 Detailed descriptions about the intelligent identification of TGEI using NRSD | 第24-31页 |
| 1.3.1 Forested landslide detection in bedrock-covered mountains with steep and rugged terrain | 第25-26页 |
| 1.3.2 LCM in CSMAL | 第26-28页 |
| 1.3.3 LCC in GDES of inland arid regions | 第28-31页 |
| 1.4 Research objectives | 第31页 |
| 1.5 The key scientific issues | 第31-32页 |
| 1.6 Structure of the dissertation | 第32-36页 |
| Chapter 2 Study areas | 第36-40页 |
| 2.1 Introduction | 第36页 |
| 2.2 The Three Gorges region of central China for forested landslide detection | 第36-37页 |
| 2.3 Wuhan City of central China for LCM in CSMAL | 第37-38页 |
| 2.4 Dunhuang Basin of northwestern China for LCC in GDEs of inland arid region | 第38-40页 |
| Chapter 3 Data and methods | 第40-61页 |
| 3.1 Data sources and pre-processing | 第40-42页 |
| 3.1.1 Airborne LiDAR data | 第40-41页 |
| 3.1.2 ZY-3 stereo satellite imagery | 第41-42页 |
| 3.1.3 RapidEye satellite imagery | 第42页 |
| 3.2 Feature calculation | 第42-54页 |
| 3.2.1 Pixel-based LiDAR derivatives | 第42-48页 |
| 3.2.2 Image segmentation and object-based LiDAR derivatives | 第48-52页 |
| 3.2.3 ZY-3 image features | 第52-53页 |
| 3.2.4 RapidEye image features | 第53-54页 |
| 3.3 Classification scheme development | 第54-56页 |
| 3.3.1 Classification scheme for landslide identification | 第54页 |
| 3.3.2 Two-level classification schemes for LCC in CSMAL | 第54-55页 |
| 3.3.3 Classification scheme for LCC in GDEs of inland arid region | 第55-56页 |
| 3.4 Training set acquisition | 第56页 |
| 3.5 Feature reduction | 第56-58页 |
| 3.5.1 Feature selection methods | 第57页 |
| 3.5.2 Feature extraction methods | 第57-58页 |
| 3.6 Classification model development and parameter optimization | 第58-59页 |
| 3.7 Test set acquisition | 第59页 |
| 3.8 Classification accuracy assessment | 第59页 |
| 3.9 Landslide recognition and assessment | 第59-61页 |
| 3.9.1 Pixel-based landslide recognition and assessment | 第59-60页 |
| 3.9.2 Object-based landslide recognition and assessment | 第60-61页 |
| Chapter 4 Forested landslide detection in the Three Gorges region | 第61-76页 |
| 4.1 Pixel-based forested landslide detection | 第61-67页 |
| 4.1.1 The utilized methods | 第61-62页 |
| 4.1.2 Acquisition of feature subset | 第62页 |
| 4.1.3 Classification accuracy assessment and analysis | 第62-64页 |
| 4.1.4 Accuracy assessment and analysis of the cross training and classification | 第64-65页 |
| 4.1.5 Delineation of landslide boundaries | 第65页 |
| 4.1.6 Conclusions | 第65-67页 |
| 4.2 Object-based forested landslide detection | 第67-74页 |
| 4.2.1 The utilized methods | 第67-68页 |
| 4.2.2 Image segmentation result | 第68-70页 |
| 4.2.3 Acquisition of object feature subset | 第70-71页 |
| 4.2.4 Classification accuracy assessment | 第71-72页 |
| 4.2.5 Landslide inventory map and accuracy assessment | 第72-74页 |
| 4.2.6 Conclusions | 第74页 |
| 4.3 Brief summary | 第74-76页 |
| Chapter 5 LCC in a complex mixed surfaced-mined and agriculturallandscape | 第76-97页 |
| 5.1 The utilized methods | 第76-78页 |
| 5.2 Results | 第78-89页 |
| 5.2.1 Acquisition of feature subset | 第78-80页 |
| 5.2.2 Parameter optimization of MLAs | 第80-81页 |
| 5.2.3 Evaluation and comparative analysis of the classification results | 第81-89页 |
| 5.3 Discussion | 第89-94页 |
| 5.3.1 The effectiveness of the employed features | 第89-90页 |
| 5.3.2 Influences of sampling design for training sets, FS method, and the size of test set, and comparison of MLAs | 第90-94页 |
| 5.4 Conclusions | 第94-97页 |
| Chapter 6 Land cover classification in an arid region | 第97-110页 |
| 6.1 LCC in inland arid region and effects of red-edge band and vegetation indices | 第97-102页 |
| 6.1.1 The utilized methods | 第97页 |
| 6.1.2 The result of LCC | 第97-98页 |
| 6.1.3 Effects of red-edge band and vegetation indices | 第98-102页 |
| 6.1.4 Conclusions | 第102页 |
| 6.2 Comparison and integration of feature selection and feature extraction methods for LCC | 第102-110页 |
| 6.2.1 The utilized methods | 第102-104页 |
| 6.2.2 Results of feature selection and feature extraction | 第104-105页 |
| 6.2.3 Comparison of feature reduction methods | 第105-107页 |
| 6.2.4 Integration of feature selection and feature extraction methods | 第107-108页 |
| 6.2.5 Conclusions | 第108-110页 |
| Chapter 7 Concluding remarks | 第110-112页 |
| 7.1 Conclusions | 第110-111页 |
| 7.2 Future work | 第111-112页 |
| Acknowledgements | 第112-113页 |
| References | 第113-129页 |
