Dimensionality Reduction of Hyperspectral Imagery

Preface
Contents
List of Figures
List of Tables
Abbreviations
Chapter 1: Introduction
	1.1 Remote Sensing
		1.1.1 Electromagnetic Spectrum
		1.1.2 Atmospheric Transmission Window
		1.1.3 Atmospheric Scattering
		1.1.4 Surface Reflection
		1.1.5 Reflectance Curve
		1.1.6 RS Data Characteristics
	1.2 Overview of Hyperspectral Remote Sensing
		1.2.1 Hyperspectral Image
		1.2.2 Multispectral Versus Hyperspectral RS
		1.2.3 Hyperspectral Sensors
		1.2.4 Summary
	1.3 Digital Image Processing
		1.3.1 Image Acquisition
		1.3.2 Image Enhancement
		1.3.3 Image Restoration
		1.3.4 Image Compression
		1.3.5 Image Segmentation
		1.3.6 Image Recognition
	1.4 Concept of Machine Learning
		1.4.1 Supervised Learning
		1.4.2 Unsupervised Learning
	References
Chapter 2: Dimensionality Reduction: State of the Art
	2.1 Feature Extraction
	2.2 Feature Selection
	References
Chapter 3: Performance Assessment and Dataset Description
	3.1 Performance Assessment
	3.2 Dataset Description
		3.2.1 Kennedy Space Center (KSC) Data
		3.2.2 Botswana Data
Chapter 4: Spectral Feature Extraction Using Pooling
	4.1 Introduction
	4.2 Methodology
	4.3 Result and Discussion
	4.4 Conclusions
	References
Chapter 5: Ranking-Based Band Selection Using Correlation and Variance Measure
	5.1 Introduction
	5.2 Methodology
		5.2.1 Reduction of Spatial Dimension and Noise
		5.2.2 Band Elimination
	5.3 Results and Discussion
	5.4 Conclusions
	References
Chapter 6: Dimensionality Reduction Using Band Optimisation
	6.1 Introduction
	6.2 Band Optimisation Using GA
		6.2.1 Introduction
		6.2.2 The General GA Technique
		6.2.3 GA-Based Band Selection Using Cluster Mean and Divergence Measure
			6.2.3.1 Introduction
			6.2.3.2 Methodology
			6.2.3.3 Result and Discussion
			6.2.3.4 Conclusions
		6.2.4 GA-Based Band Selection Using Wavelet and Signal Entropy
			6.2.4.1 Introduction
			6.2.4.2 Methodology
			6.2.4.3 Results and Discussion
			6.2.4.4 Conclusions
	6.3 Band Optimisation Using PSO
		6.3.1 Introduction
		6.3.2 The General PSO Technique
		6.3.3 Methodology
			6.3.3.1 Image Filtering
			6.3.3.2 Generation of Unique Patterns
			6.3.3.3 BS Using Modified PSO
		6.3.4 Results and Discussion
		6.3.5 Conclusions
	6.4 Conclusions
	References
Chapter 7: Data-Driven Approach for Hyperspectral Band Selection
	7.1 Introduction
	7.2 Unsupervised Data-Driven Band Selection Approach
		7.2.1 Introduction
		7.2.2 Methodology
			7.2.2.1 Multifeature Analysis
			7.2.2.2 Band Selection
		7.2.3 Result and Discussion
		7.2.4 Conclusions
	7.3 Supervised Data-Driven Band Selection Approach
		7.3.1 Introduction
		7.3.2 Methodology
			7.3.2.1 Sample Pattern Extraction for Each Class
			7.3.2.2 Quantisation-Based Band Selection
			7.3.2.3 Correlation Coefficient and Variance Measure–Based Redundancy Removal
		7.3.3 Result and Discussion
		7.3.4 Conclusions
	7.4 Conclusions
	References
Chapter 8: Concluding Remarks and Way Forward
	8.1 Concluding Remarks
	8.2 Way Forward
	References
Index