Abbreviations
Chapter 1 Introduction
1.1 Overview on Remote Sensing
1.1.1.The Electromagnetic Spectrum
1.1.2.Main Modes of Observation
1.2Remote Sensing Images
1.2.1Hyperspectral Images
1.2.2VHR Multispectral Remote Sensing Images
1.3Overview of Remote Sensing Image Classification
1.4Objectives of the thesis
1.5 Main Contributions
1.6 Structure of the Thesis
Chapter 2 Related Works and Background
2.1 Principal Concept of Spectral-Spatial Feature Extraction Techniques
2.2 Recent Spectral-Spatial Feature Extraction Methods
2.2.1Morphological profiles and their extended versions for spatial feature extraction
2.2.2 Deep learning approaches for spectral-spatial image classification
2.2.3 Texture feature extraction techniques
2.3Support Vector Machine for Remote Sensing Images Classification
2.3.1 Fundamental Concept of SVM
2.3.2 Multiclass classification based on SVM
Chapter 3 Feature Extraction Using EMAP-SAE and Spectral-EMAP-SAE for RS-Image Classification
3.1 Introduction
3.2 Spatial Feature Extraction Based on Extended Multi-Attribute Profiles
3.3 Sparse Autoencoder for Feature Extraction and Dimensionality Reduction
3.4 EMAP-SAE and Spectral-EMAP-SAE for Hyperspectral and Multispectral Image Classification
3.4.1 Description of EMAP-SA and Spectral-EMAP-SAE Frameworks
3.4.2 Description of Remote Sensing Data in Investigation
3.4.3 Experiments and Discussion
3.5 Summary
Chapter 4 D-SS Frame Based on Feature Extraction and Fusion for Panchromatic and Multispectral Image Classification
4.1 Introduction
4.2 Feature Extraction Techniques for Panchromatic and Multispectral Images
4.2.1Sparse Autoencoder and Deep Sparse Autoencoder for Spectral-Spatial Feature Extraction
4.2.2 MS and PAN Feature-Level Fusion
4.2.3 General Process of Spectral-Spatial Classification of PAN and MS Images
4.3 Experimental Results
4.3.1 Datasets Description
4.3.2 Experiments and Analysis
4.4 Summary
Chapter 5 Multiple Spectral-Spatial Feature Extraction and Fusion Based on RS-MSSF Frame
5.1 Introduction
5.2 Feature Extraction and Fusion Scheme
5.3 Feature Extraction Techniques
5.3.1 SSAE for Spectral Features and EMAP for Shape Features
5.3.2 Texture Feature Extraction Based on FGLCM
5.3.3 Multiple Features Fusion
5.4 Experiments Demonstration
5.4.1 Datasets in Investigation
5.4.2 Parameter Tuning and Classification Performance
5.5 Summary
Conclusions
参考文献
List of Publications
声明
致谢
Resume