book

Spaceborne Synthetic Aperture Radar Remote Sensing

Name: Spaceborne Synthetic Aperture Radar Remote Sensing
ISBN: 9781000803167

by Shashi Kumar, Paul Siqueira, Himanshu Govil, Shefali Agrawal

March 2023

Intermediate to advanced

432 pages

17h 24m

English

CRC Press

Read now

Unlock full access

Cover
Half Title
Title Page
Copyright Page
Table of Contents
Preface
Editor Biographies
Contributors
1 Synthetic Aperture Radar Remote Sensing
1.1 Introduction
1.1.1 Coherent Radar
1.2 Synthetic Aperture Radar (SAR) Geometry

1.3 Normalized Radar Backscatter Cross-Section
1.4 Local Angle of Incidence
1.5 SAR Polarizations
1.6 SAR Imaging Mode
References
2 Speckle Reduction in SAR Images
2.1 Introduction
2.2 SAR Imaging Modalities
2.2.1 Amplitude/Intensity Modality
2.2.2 Polarimetric Modality
2.2.3 Multi-Frequency Modality2.2.4 Interferometric Modality2.2.5 Polarimetric-Interferometric Modality2.3 Single-Channel SAR Images Restoration2.3.1 Speckle Model
2.3.2 Despeckling Filters (1/2)
2.3.2.1 Local Framework2.3.2.2 Nonlocal Framework
2.3.2 Despeckling Filters (2/2)
2.3.2.3 Deep Learning Strategies for SAR Image Despeckling
2.3.3 Despeckling Evaluation Metrics
2.3.3.1 Qualitative Assessments2.3.3.2 Quantitative Assessments
2.4 Multi-Channel SAR Images Restoration
2.4.1 Speckle Model
2.4.2 Despeckling Filters
2.4.2.1 Local Framework2.4.2.2 Nonlocal (NL) Framework2.4.2.3 Deep Learning Framework
2.4.3 Despeckling Evaluation Metrics
2.5 Open Issues and Future Trends
Note
References
3 Polarimetric Interferometric Decomposition
3.1 Introduction
3.2 Limitations in Target Identification
3.3 Polarimetric Interferometric Fundamentals and Advancements
3.3.1 Radar Imaging
3.3.2 Polarization of EM Waves
3.3.3 Scattering Matrix
3.3.4 PolSAR
3.3.5 Scattering Vectors, Coherency Matrix, and Covariance Matrix
3.3.6 Target Decomposition
3.3.7 Model-Based Decomposition
3.3.8 Classical Three-Component Decomposition Model
3.3.9 Four-Component Decomposition Model3.3.10 Polarization Orientation Angle (POA) Compensation
3.4 Polarimetric SAR Interferometry (PolInSAR)
3.4.1 Polarimetric Interferometric Decomposition Model
3.4.2 Hybrid Decomposition3.4.3 Data Preprocessing
3.4.4 Fine Co-Registration
3.4.5 PolInSAR Coherence Estimation
3.4.6 Coherence Optimization and Selection
3.4.7 Entropy and Alpha Angle
3.4.8 Model Workflow and the Constraint Equation
3.5 Experiment With Spaceborne Data
3.5.1 Coherence Selection
3.5.2 Coherence Optimization
3.5.3 Scattering Ambiguity in Generic Decomposition Models
3.6 Constraint Equation
3.7 Results of Hybrid Decomposition Model
3.7.1 Comparative Analysis
3.7.2 Comparison With Existing Decomposition Models (1/2)
3.7.2 Comparison With Existing Decomposition Models (2/2)
3.7.3 Scattering Entropy Evaluation
3.7.4 Limitation of POA Compensation
3.8 Discussions
3.8.1 Factors Affecting PolInSAR Coherence
3.8.2 PolInSAR Coherence and Decomposition Modeling
3.8.3 Statistical Significance of PolInSAR Coherence
3.9 Conclusion and Recommendations
References
4 Implementation of Machine Learning Classification Models On Multifrequency Band SAR Dataset
4.1 Introduction
4.2 Study Area
4.3 Methodology
4.4 Pre-Processing
4.5 PolSAR Decomposition
4.6 Training Data Generation
4.7 Methods for Machine Learning Classifiers
4.7.1 K-Nearest Neighbor (KNN) Classification Model
4.7.2 Naïve Bayes Classification
4.7.3 Random Forest Classification
4.7.4 Support Vector Machine Classification
4.8 Conclusion and Recommendations
References
5 Implementation of Neural Network-Based Classification Models On Multifrequency Band SAR Dataset
5.1 Introduction
5.2 Neural Network Models
5.3 Multi-Layer Perceptron (MLP) Classification Model5.3.1 Study Area
5.3.2 Methodology
5.3.3 Pre-Processing
5.3.4 PolSAR Decomposition
5.3.5 Training Data Generation
5.3.6 Architecture of Multi-Layer Perceptron (MLP) Classification Model
5.4 Compact CNN Classification Model Using Sliding-Window Operations
5.5 The DeepLabv3+ Classification Model
5.6 Conclusion and Recommendations
References
6 Improved Data Fusion-Based Land Use/Land Cover Classification Using PolSAR and Optical Remotely Sensed Satellite Data: ...
6.1 Introduction
6.1.1 Problems in Monitoring Land Use/Land Cover Change
6.1.2 SAR Polarimetry
6.1.3 Application of PolSAR in Land Cover (LC) Classification
6.1.4 Advantages of PolSAR Over Optical Datasets
6.2 Materials and Methods6.2.1 Study Area and Datasets6.2.1.1 Study Area6.2.1.2 Datasets
6.2.2 Polarimetric Decomposition
6.2.2.1 Coherent Decompositions6.2.2.2 Incoherent Decompositions
6.2.3 Data Fusion Techniques Used
6.2.3.1 Principal Component Analysis (PCA)6.2.3.2 Wavelet-Based Fusion6.2.3.3 Intensity Hue Saturation Method6.2.3.4 High Pass Filtering (HPF) Fusion
6.2.3.5 Average Method
6.2.4 Selection of Classifiers6.2.4.1 Parametric Classifiers6.2.4.2 Non-Parametric Classifiers6.2.4.3 K-Means Nearest Neighbourhood Classifier (KNN)
6.2.4.4 Support Vector Machine (SVM) Classification
6.2.4.5 Artificial Neural Network (ANN) Classifier6.2.4.6 Random Forest Classifier6.2.4.7 Object-Based Image Classification (OBIC)6.2.5 Methods
6.3 Results and Conclusion
6.3.1 Polarimetric Decomposition6.3.2 Classification Results6.3.2.1 KNN Classification6.3.2.2 Random Forest (RF) Classification6.3.2.3 Artificial Neural Network (ANN) Classification6.2.4.4 Support Vector Machine (SVM) Classification
6.2.4.5 Object-Based Image Classification (OBIC)
6.4 Conclusions
References (1/2)
References (2/2)
7 Polarimetric SAR Descriptors for Rice Monitoring
7.1 Introduction
7.2 Methodology
7.2.1 The Kennaugh Matrix Framework
7.2.2 Target Characterization Parameters
7.2.3 Scattering Power Decomposition
7.3 Study Area and Satellite Data
7.3.1 Indian Test Site7.3.1.1 Satellite Dataset and Pre-Processing7.3.2 Spanish Test Site7.3.2.1 Satellite Dataset and Pre-Processing
7.3.3 In-Situ Measurement Procedures and Rice Morphology
7.3.3.1 Morphological Characteristics of Rice Across Its Phenological Stages
7.4 Results and Discussion
7.4.1 Temporal Analysis Over Two Test Sites and Different Incidence Angle
7.4.2 Rice Phenology Classification
7.5 Conclusions
Acknowledgments
References
8 Synergistic Fusion of Spaceborne Polarimetric SAR and Hyperspectral Data for Land Cover Classification
8.1 Introduction
8.1.1 Background
8.2 Study Area and Datasets Used
8.2.1 Study Area8.2.2 Datasets Used8.2.2.1 EO-1 Hyperion Data8.2.2.2 Radarsat-2 Data8.2.2.3 ALOS PALSAR Data
8.3 Research Methodology
8.3.1 Processing of the Datasets
8.3.1.1 Pre-Processing of Hyperion Data8.3.1.2 Processing of PolSAR Data8.3.1.3 Multiple-Component Scattering Model (MCSM) Decomposition8.3.1.4 Surface Scattering8.3.1.5 Double-Bounce Scattering8.3.1.6 Volume Scattering8.3.1.7 Helix Scattering8.3.1.8 Wire Scattering
8.3.1.9 Geocoding of Extracted Polsar Parameters and Co- Registration With Hyperspectral Image
8.3.2 Pixel-Level Fusion8.3.2.1 High-Pass Filter Fusion
8.3.2.2 Wavelet Fusion
8.3.2.3 Gram–Schmidt Fusion8.3.3 Feature-Level Fusion8.3.3.1 Kernel-Based Principal Component Analysis Feature Extraction in Hyperion
8.3.3.2 MCSM-Based Feature Extraction in Radarsat-2 and ALOS PALSAR
8.3.4 Decision-Level Fusion8.3.4.1 One-Against-All Multiclass SVM Classification
8.4 Classification and Accuracy Assessment
8.4.1 Classification8.4.1.1 Support Vector Machines (SVM)
8.4.2 Accuracy Assessment
8.4.2.1 Cross-Validation8.4.2.2 Holdout Method With Parameter Tuning
8.4.3 Comparative Analysis
8.5 Results and Discussion8.5.1 Results8.5.1.1 Results of Pixel-Level Fusion
8.5.1.2 Results of Feature-Level Fusion
8.5.1.3 Results of Decision-Level Fusion
8.5.2 Comparative Analysis of All The Three Levels of Fusion of Hyperspectral and PolSAR Data
8.6 Conclusion and Recommendation
8.6.1 Conclusion
8.6.2 Recommendations
References
9 Marine Oil Slick Detection Using Synthetic Aperture Radar Remote Sensing Techniques
9.1 Introduction
9.2 Types of Oil
9.3 Weathering Processes
9.4 Major Oil Spill Incidents
9.5 Cleanup Procedure and Associated Costs
9.5.1 Monetary Costs Associated With Oil Spills
9.6 Oil Spill Detection Procedure
9.7 SAR Remote Sensing for Oil Spill Detection
9.7.1 Polarization Signature Analysis of the Oil Spill Patches
9.7.2 Decomposition Models
9.7.2.1 Coherent Decomposition Models9.7.2.2 Incoherent Decomposition Models9.7.2.3 Eigenvalue/Eigenvector Decomposition – Decomposition
9.8 Separability Analysis
9.9 Classification Techniques
9.9.1 Support Vector Machine
9.9.2 Wishart Supervised Classification
9.10 Accuracy Assessment
9.10.1 AUC (Area Under the Curve) and ROC (Receiver Operating Characteristics)
9.11 Conclusions and Future Work
References
10 Spaceborne C-Band PolSAR Backscatter and PolInSAR Coherence-Based Modeling for Forest Aboveground Biomass Estimation
10.1 A Brief History of Applications of Synthetic Aperture Radar (SAR) for Biophysical Parameter Retrieval
10.2 Doon Valley Forests at a Glance
10.3 Overview of the Fundamentals and Methodological Development
10.4 Illustration of the Outcomes of the Analysis
10.5 Inferences
References
11 Analysis of Polarimetric Techniques for Characterization of Glacial Feature
11.1 Introduction
11.2 Study Area and Datasets
11.2.1 Study Area
11.2.1.1 Gangotri Glacier
11.2.1.2 Siachen Glacier11.2.2 Datasets11.4 Methodology11.4.1 Calculation of Backscattering Coefficient σ0
11.4.2 Scattering Matrix
11.4.3 Target Decomposition
11.4.3.1 Eigenvalue–Eigenvector Decomposition11.4.3.2 Freeman–Durden Decomposition
11.4.4 Stokes Parameters
11.5 Results and Discussion11.5.1 Backscattering Coefficient σo Image Interpretation11.5.1.1 Siachen Glacier Ablation Zone11.5.1.2 Gangotri Glacier
11.5.2 Results of H-A-Α Decomposition
11.5.2.1 Gangotri Glacier
11.5.2.2 Siachen Glacier
11.5.3 Results of Freeman–Durden Decomposition11.5.3.1 Gangotri Glacier
11.5.3.2 Siachen Glacier
11.5.4 Stokes Parameter-Based Approach11.5.4.1 Circular Polarization Ratio (CPR)
11.6 Conclusion
References
12 Spaceborne SAR Application to Study Ice Flow Variation of Potsdam Glacier and Polar Record Glacier, East Antarctica
12.1 Introduction
12.2 Role of Spaceborne SAR in Antarctica
12.3 Glacier Velocity Estimation Using Spaceborne SAR
12.4 Study Area and Dataset
12.5 Methodology
12.5.1 Glacier Velocity Estimation Using DInSAR
12.5.2 Glacier Velocity Estimation Using Offset Tracking
12.5.3 Field Measurements
12.6 Results and Discussion
12.6.1 DInSAR-Based Glacier Velocity Estimation for Potsdam Glacier
12.6.2 Offset Tracking Based Glacier Velocity Estimation for PRG
12.7 Conclusion
References
13 Multi-Temporal SAR Interferometry: Theory, Processing, and Applications
13.1 Introduction
13.2 Synthetic Aperture Radar (SAR)
13.3 Geometric Distortions in SAR
13.4 SAR Interferometry (InSAR)
13.5 MT-InSAR (Advanced InSAR Technique) Processing Chain
13.5.1 StaMPS: Stanford Method for Persistent Scatterers13.5.1.1 Generation of Interferogram13.5.1.2 PS Selection13.5.1.3 Phase Stability Estimation13.5.1.4 Phase Unwrapping and Displacement Estimation13.5.1.5 SBAS
13.6 Monitoring of Landslides in Nainital (Uttarakhand, India) Using StaMPS
13.7 Monitoring of Landslides in Nainital (Uttarakhand, India) Using StaMPS
13.8 Surface Deformation Measurement of the L’Aquila Region
13.9 Conclusions
References
14 SAR for Cultural Heritage Monitoring
14.1 Remote Sensing for Cultural Heritage Sites
14.2 Study Area
14.2.1 Angkor Wat
14.2.2 Structure
14.2.3 Hydraulic Network
14.2.4 Geomorphological
14.3 Dataset
14.4 PSInSAR Processing of Sentinel Data
14.5 Results
14.5.1 PS Identification14.5.2 LOS Displacement of 2006 to 2009
14.5.3 LOS Displacement of 2014 to 2018
14.5.4 LOS Displacement 2017 to 2021
14.6 Discussion
14.7 Conclusion
References
15 Extraction and Evaluation of Lineaments From DEMs Generated From Different Bands of Microwave Data and Optical Data: …
15.1 Introduction
15.2 Study Area
15.2.1 Geological Settings of the Study Area
15.3 Data Used
15.3.1 Microwave Data15.3.1.1 Sentinel-1A15.3.1.2 ALOS-I PALSAR
15.3.2 Optical Data
15.3.2.1 ALOS World 3D-30m (AW3D30)15.4 Methodology15.4.1 Generation of DEM15.4.1.1 Sentinel-1A
15.4.1.2 ALOS-I PALSAR
15.4.2 Comparison of DEMs15.4.3 Automatic Lineament Extraction
15.4.4 Comparison of Extracted Lineaments
15.5 Results
15.6 Conclusions
AcknowledgmentsReferences
16 Scatterer-Based Deformation Monitoring Induced Due to Coal Mining By DInSAR Techniques
16.1 Introduction
16.2 Study Area and Data Used
16.2.1 Study Area16.2.2 Datasets Used
16.3 Methodology
16.3.1 Baseline Estimation
16.3.2 Co-Registration
16.3.3 Generation of Interferogram
16.3.4 Coherence Estimation
16.3.5 Topographic Phase Removal
16.3.6 Phase Unwrapping
16.3.7 Phase Filtering
16.3.8 Identification and Selection of Persistent Scatterers16.3.9 Atmospheric Phase Screen Estimation and Its Correction
16.3.10 Persistent Scatterer Coherence Map Generation
16.3.11 Generation of Cumulative Displacement Map and Monitoring Deformation History16.4 Results and Discussion
16.5 Conclusion
References
17 An Insight to the Lunar Surface: Characterization From the L- and S-Band Polarimetric Data
17.1 Introduction
17.2 The Moon: A Catapult for Future Space Missions
17.2.1 Importance of the Moon in the Indian Space Program
17.2.2 Human Footprints On the Moon: The Manned Missions On the Lunar Surface17.2.3 The Chronicle of Lunar Exploration: Lander, Orbiter, and Impactor
17.2.4 Future Lunar Missions
17.3 The New Era of SAR-Based Missions for the Lunar Surface
17.3.1 Chandrayaan-117.3.2 Lunar Reconnaissance Orbiter17.3.3 Chandrayaan-2
17.4 Crater Formation
17.4.1 Types of Craters
17.5 Lunar Pyroclastic Deposits
17.6 Lunar Rilles
17.7 South Pole
17.8 Circular Polarization Ratio
17.9 Decomposition Techniques
17.9.1 M-δ Decomposition
17.9.2 M-χ Decomposition
17.9.3 M-α Decomposition
17.10 Study Area and Datasets
17.10.1 Shackleton Crater17.10.2 Erlanger Crater17.10.3 Slater Crater
17.10.4 Rimae Sulpicius Gallus
17.11 Results
17.11.1 Shackleton Crater17.11.1.1 M-δ (Delta) Decomposition17.11.1.2 M-Χ (Chi) Decomposition17.11.1.3 M-α (Alpha) Decomposition
17.11.2 Analysis of CPR
17.11.3 Slater Crater
17.11.3.1 Barnes Decomposition
17.11.4 Erlanger Crater
17.11.4.1 M-δ (Delta) Decomposition17.11.4.2 M-Χ (Chi) Decomposition17.11.4.3 M-α (Alpha) Decomposition17.11.4.4 Analysis of CPR17.11.4.5 Barnes Decomposition
17.11.5 Rimae Sulpicius Gallus
17.12 Conclusions
References
18 Synthetic Aperture Radar (SAR) Data Calibration and Validation
18.1 Introduction
18.2 Significance of SAR Calibration
18.3 Types of SAR Calibrations
18.3.1 Internal Calibration
18.3.2 External Calibration
18.3.2.1 Active Targets18.3.2.2 Passive Targets18.3.2.3 Distributed Targets
18.4 Corner Reflectors
18.4.1 Types of Corner Reflectors18.4.1.1 Trihedral Corner Reflectors18.4.1.2 Corner Reflectors for Polarimetric Calibration18.4.1.3 Dihedral Corner Reflectors18.4.1.4 Polarization Selective Dihedrals18.4.1.5 Interferometric SAR Calibration
18.4.1.6 Self-Illuminating Corner Reflectors
18.4.2 Construction and Fabrication of Corner Reflectors
18.4.3 Selection of Calibration Site and Deployment of Corner Reflectors
18.5 Types of SAR Calibration and Validation
18.5.1 Radiometric Calibration
18.5.2 Geometric Calibration
Appendix
References
Index (1/2)
Index (2/2)

Content preview from Spaceborne Synthetic Aperture Radar Remote Sensing

119

Neural Network-Based Classiﬁcation Models

119

5.6 CONCLUSION AND RECOMMENDATIONS

The outcomes of this study are aligned with previous research. The neural network models have

been observed to be overwhelmingly good for LULC classication. In this chapter, three different

NN models – MLP, compact CNN, and DeepLabv3+ – have been implemented on different SAR

datasets. The MLP classier is the basic NN model and has been applied on the TerraSAR- X,

RADARSAT- 2, GF- 3, and ALOS- 2/ PALSAR- 2 datasets; and the MLP is observed to work better

for the TerraSAR- X dataset. The compact CNN model is implemented on the AIRSAR L- band and

RADARSA

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781000803167

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Spaceborne Synthetic Aperture Radar Remote Sensing

by Shashi Kumar, Paul Siqueira, Himanshu Govil, Shefali Agrawal

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

More than 5,000 organizations count on O’Reilly

Julian F.

Addison B.

Amir M.

Mark W.

You might also like

Multi-Antenna Synthetic Aperture Radar

Synthetic Aperture Radar Polarimetry

Multi-Dimensional Imaging with Synthetic Aperture Radar

Inverse Synthetic Aperture Radar Imaging With MATLAB Algorithms, 2nd Edition

Publisher Resources