Multi-resolution Image Fusion in Remote Sensing

✍ Scribed by Manjunath V. Joshi, Kishor P. Upla

• Publisher: Cambridge University Press
• Year: 2019
• Tongue: English
• Leaves: 255
• Category: Library

✦ Synopsis

Written in an easy-to-follow approach, the text will help the readers to understand the techniques and applications of image fusion for remotely sensed multi-spectral images. It covers important multi-resolution fusion concepts along with the state-of-the-art methods including super resolution and multi stage guided filters. It includes in depth analysis on degradation estimation, Gabor Prior and Markov Random Field (MRF) Prior. Concepts such as guided filter and difference of Gaussian are discussed comprehensively. Novel techniques in multi-resolution fusion by making use of regularization are explained in detail. It also includes different quality assessment measures used in testing the quality of fusion. Real-life applications and plenty of multi-resolution images are provided in the text for enhanced learning.

✦ Table of Contents

Contents
List of Figures
List of Tables
Preface
Acknowledgments
1 Introduction
1.1 Characteristics of Remotely Sensed Imagery
1.1.1 Multi-spectral images
1.1.2 Panchromatic image
1.1.3 Hyper-spectral images
1.2 Low Spatial Resolution Imaging
1.3 Image Fusion in Remotely Sensed Images
1.4 Multi-resolution Image Fusion: An Ill-posed Inverse Problem
1.5 Indian Remote Sensing Satellites
1.6 Applications of Image Fusion
1.7 Motivation
1.8 Organization of the Book
2 Literature Review
2.1 Projection Substitution Based Techniques
2.2 Multi-resolution Based Techniques
2.3 Model Based Fusion Approaches
2.4 Hyper-spectral Sharpening Methods
2.5 Conclusion
3 Image Fusion Using Different Edge-preserving Filters
3.1 Related Work
3.2 Fusion Using Multistage Guided Filter (MGF)
3.2.1 Multistage guided filter (MGF)
3.2.2 Proposed approach using guided filter
3.3 Fusion Approach Using Difference of Gaussians (DoGs)
3.3.1 Difference of Gaussians (DoGs)
3.3.2 Proposed approach using DoGs
3.4 Experimental Illustrations
3.4.1 Experimentations: Ikonos-2 dataset
3.4.2 Experimentations: Quickbird dataset
3.4.3 Experimentations: Worldview-2 dataset
3.4.4 Computational complexity
3.5 Conclusion
4 Image Fusion: Model Based Approach with Degradation Estimation
4.1 Previous Works
4.2 Description of the Proposed Approach Using Block Schematic
4.3 Background: Contourlet Transform (CT)
4.4 Contourlet Transform Based Initial Approximation
4.5 Forward Model and Degradation Estimation
4.6 MRF Prior Model
4.7 MAP Estimation and Optimization Process
4.7.1 MAP estimation
4.7.2 Optimization process
4.8 Experimentations
4.8.1 Effect of decimation matrix coefficients on fusion
4.8.2 Effect of MRF parameter γm on fusion
4.8.3 Fusion results for degraded dataset: Ikonos-2
4.8.4 Fusion results for degraded dataset: Quickbird
4.8.5 Fusion results for degraded dataset: Worldview-2
4.8.6 Fusion results for un-degraded (original) datasets: Ikonos-2, Quickbird and Worldview-2
4.8.7 Spectral distortion at edge pixels
4.8.8 Computational time
4.9 Conclusion
5 Use of Self-similarity and Gabor Prior
5.1 Related Work
5.2 Block Schematic of the Proposed Method
5.3 Initial HR Approximation
5.4 LR MS Image Formation Model and Degradation Matrix Estimation
5.5 Regularization Using Gabor and MRF Priors
5.5.1 Optimization process
5.6 Experimental Results
5.6.1 Experimental setup
5.6.2 Experimental results on degraded and un-degraded Ikonos-2 datasets
5.6.3 Experimental results on degraded and un-degraded Quickbird datasets
5.6.4 Experimental results on degraded and un-degraded Worldview-2 datasets
5.6.5 Comparison of fusion results with CS and TV based approaches
5.6.6 Computation complexity
5.7 Conclusion
6 Image Fusion: Application to Super-resolution of Natural Images
6.1 Related Work
6.2 Estimation of Close Approximation of the SR Image
6.3 Refining SR Using MAP–MRF Framework
6.4 MRF Prior and SR Regularization
6.4.1 Optimization process
6.5 Experimental Demonstrations
6.5.1 SR results on gray scale images
6.5.2 SR results on color images
6.6 Conclusion
7 Conclusion and Directions for Future Research
7.1 Conclusion
7.2 Future Research Work
Bibliography