Non-Linear Spectral Unmixing of Hyperspectral Data / Somdatta Chakravortty.

Format:

Book

Author/Creator:

Chakravortty, Somdatta, author.

Language:

English

Subjects (All):

Hyperspectral imaging.

Artificial satellites in remote sensing.

Image processing--Digital techniques.

Image processing.

Nonlinear optics.

Optical detectors.

Mangrove forests--Sundarbans (Bangladesh and India)--Remote sensing--Case studies.

Mangrove forests.

Physical Description:

1 online resource (131 pages)

Edition:

First edition.

Place of Publication:

Boca Raton, FL : CRC Press, [2025]

Summary:

This book is based on satellite image processing focussing on the potential of hyperspectral image processing research taking a case study-based approach. It covers the background, objectives, and practical issues related to HIP and discusses the needs/potentials of said technology for discrimination of pure and mixed endmembers in pixels.

Contents:

Cover

Half Title

Title Page

Copyright Page

Table of Contents

Preface

About the Author

1 Introduction

1.1 Hyperspectral Image Processing

1.2 Availability of Hyperspectral Data

1.3 The Objective of the Book

1.4 Content and Organization of the Book

1.5 Societal Relevance

1.6 A Case Study of Mangrove Endmembers for Result Analysis

2 Hyperspectral Image Processing: A Review

2.1 Existing Multispectral Technology and Its Limitations

2.2 Potentials of Hyperspectral Remote Sensing

2.3 Automated Endmember Detection

2.4 Spectral Mixture Analysis

2.4.1 Linear Spectral Unmixing On Hyperspectral Data

2.4.2 Non-Linear Spectral Unmixing of Hyperspectral Data

3 Preprocessing of Data

3.1 Preprocessing of Data

3.2 Ground Survey

3.2.1 Case Study-Based Ground Truth Results

3.3 Spectral Library Development From Ground Data

3.3.1 Case Study-Based Results

3.4 Preprocessing of Hyperspectral Data

3.4.1 Acquisition of Hyperspectral Data

3.4.2 Bands and Spectral Ranges for Hyperspectral Data

3.4.3 Removal of Absorption Bands and Bands Having No Information

3.4.4 Removal of Bad Columns and Vertical Stripes

3.5 Atmospheric Correction

3.5.1 Analysis of Atmospheric Correction Results

3.6 Geometric Correction

3.6.1 Image-To-Image Registration

3.6.2 Spatial Interpolation Using Coordinate Transformations

3.6.3 Intensity Interpolation and Resampling

3.6.4 Accuracy Assessment of Ground Control Points

3.6.5 Analysis of Geometric Correction Results

4 Endmember Detection

4.1 Automated Endmember Detection Algorithms

4.1.1 Automated Target Generation Process (ATGP)

4.1.2 N-FINDER

4.1.3 Pixel Purity Index (PPI)

4.2 Spectral Signature Analysis of Extracted Endmembers

4.3 Experimental Results of Analysis Between Endmember Detection Algorithms.

5 Least-Squares-Based Linear Spectral Unmixing For Pure Endmembers

5.1 Linear Spectral Unmixing

5.2 Linear Spectral Unmixing of Endmembers: Case Study of Mangrove Ecosystem

5.3 Case Study of Fractional Abundance Estimation of Mangrove Species

5.3.1 Endmember Detection

5.3.2 Fractional Abundance

5.3.3 Accuracy Assessment

6 Non-Linear Unmixing for Classification of Mixed Endmembers

6.1 Limitations of Linear Spectral Unmixing

6.2 Non-Linear Unmixing Models

6.3 Nascimento's Bilinear Spectral Unmixing Model

6.4 Fan's Bilinear Unmixing Model

6.5 Hapke's Bidirectional Model

6.6 Higher-Order Non-Linear Spectral Unmixing Models

6.7 Experimental Results of Analysis of Lower to Higher-Order Non-Linear Models With Case Studies

7 Fuzzy Logic-Based Non-Linear Spectral Unmixing

7.1 Fuzzy Logic-Based Non-Linear Spectral Unmixing

7.2 Fuzzy C-Means (FCM)

7.3 Possibilistic C-Means (PCM)

7.4 Entropy-Based Fuzzy C-Means Unmixing

7.5 Spatial Fuzzy-Based Unmixing

7.6 Applications of Fuzzy-Based Non-Linear Unmixing

8 Machine Learning Models for Classification of Hyperspectral Data

8.1 Machine Learning Models for Classification of Hyperspectral Data

8.2 Classification Models

8.3 Prediction Models

8.3.1 Deep Learning

8.4 Merits and Demerits of the Models Through Experimental Analysis

8.5 Role of Machine/Deep Learning in Spatial Big Data Analytics

Case Study: Evaluating the Performance of a Convolutional Neural Network (CNN) Model in Classifying Hyperspectral Images in the Andaman and Nicobar Islands

9 Ecodynamic Modeling

9.1 Ecodynamic Modeling

9.2 The Ecodynamic Model

9.2.1 Ecodynamic Model and Time-Series Analyses of Hyperspectral Data

9.2.2 Application of the Ecodynamic Model in Change Detection of Coastal Endmembers.

9.2.3 Application of Ecodynamic Model in Competitive Growth Rate of Coastal Endmembers

9.2.4 Application of the Ecodynamic Model in Competitive Survival Capacity of Coastal Endmembers

9.2.5 Application of the Ecodynamic Model in Pair-Wise Competition of Coastal Endmembers

9.3 Application of Lotka-Volterra Model in Competitive Growth Rate of Species

9.4 Challenges of Hyperspectral Data for Mapping and Classification - Gaps and Solutions

Bibliography

Index.

Notes:

Includes bibliographical references and index.

Description based on publisher supplied metadata and other sources.

Description based on print version record.

ISBN:

9781040112618

1040112617

9781003432623

100343262X

9781040112557

1040112552

OCLC:

1443930317