Remote Sensing Image Classification in R

✍ Scribed by Courage Kamusoko

• Publisher: Springer Nature
• Year: 2019
• Tongue: English
• Leaves: 201
• Series: Springer Geography
• Edition: 1 ed. 2019
• Category: Library

✦ Synopsis

This book offers an introduction to remotely sensed image processing and classification in R using machine learning algorithms. It also provides a concise and practical reference tutorial, which equips readers to immediately start using the software platform and R packages for image processing and classification.

This book is divided into five chapters. Chapter 1 introduces remote sensing digital image processing in R, while chapter 2 covers pre-processing. Chapter 3 focuses on image transformation, and chapter 4 addresses image classification. Lastly, chapter 5 deals with improving image classification.

R is advantageous in that it is open source software, available free of charge and includes several useful features that are not available in commercial software packages. This book benefits all undergraduate and graduate students, researchers, university teachers and other remote- sensing practitioners interested in the practical implementation of remote sensing in R.

✦ Table of Contents

Preface
Who should use this workbook?
How is this workbook organized?
Conventions used in this workbook
Data sets, R scripts, and online resources
Acknowledgements
Contents
Abbreviations and Acronyms
1 Remote Sensing Digital Image Processing in R
Abstract
1.1 Introduction
1.1.1 Remote Sensing Digital Image Processing
1.1.2 Overview of Machine Learning
1.2 Overview of R
1.2.1 What Is R?
1.2.2 Installing R
1.2.3 RStudio
1.2.4 R Packages
1.2.5 Overview of Data Type, Structure, and Functions
1.2.6 Getting Help
1.2.7 Handling Errors and Warnings
1.2.8 Other Issues
1.3 Data and Test Site
1.3.1 Landsat Imagery
1.3.2 Reference Data Sets
1.3.3 Overview of Harare Metropolitan Province
1.4 Tutorial 1: A Quick Guide to R
1.4.1 Objectives
1.4.2 Overview of the Packages
1.4.3 R Basics
1.4.4 Procedure
1.5 Summary
1.6 Additional Exercises
References
2 Pre-processing
Abstract
2.1 Background
2.1.1 Radiometric Correction
2.2 Tutorial 1: Display Landsat 5 TM Imagery
2.2.1 Objectives
2.2.2 Overview of Packages
2.2.3 Procedure
2.3 Tutorial 2: Radiometric Correction and Reprojection
2.3.1 Objectives
2.3.2 Procedure
2.4 Summary
2.5 Additional Exercises
References
3 Image Transformation
Abstract
3.1 Background
3.1.1 Spectral Indices
3.1.2 Texture Indices
3.2 Tutorial 1: Vegetation Indices
3.2.1 Objective
3.2.2 Overview of the Packages
3.2.3 Procedure
3.3 Tutorial 2: Texture Analysis
3.3.1 Objective
3.3.2 Overview of the Packages
3.3.3 Procedure
3.4 Summary
3.5 Additional Exercises
References
4 Image Classification
Abstract
4.1 Overview of Image Classification
4.1.1 k-Nearest Neighbors (KNN)
4.1.2 Artificial Neural Networks (ANN)
4.1.3 Single Decision Trees (DT)
4.1.4 Support Vector Machines (SVM)
4.1.5 Random Forest (RF)
4.2 Tutorial 1: Single Date Image Classification
4.2.1 Objectives
4.2.2 Overview of Packages
4.2.3 Procedure
4.2.4 Summary for Tutorial 1
4.3 Tutorial 2: Multidate Landsat Image Classification
4.3.1 Objective
4.3.2 Procedure
4.3.3 Summary for Tutorial 2
4.4 Summary
4.5 Additional Exercises
References
5 Improving Image Classification
Abstract
5.1 Overview
5.2 Feature Selection
5.2.1 Importance of Feature Selection
5.2.2 Recursive Feature Elimination (RFE)
5.3 Tutorial 1: Image Classification Using Multiple Data Sets
5.3.1 Objective
5.3.2 Procedure
5.3.3 Summary for Tutorial 1
5.4 Tutorial 2: Image Classification Using Multiple Data Sets with Feature Selection
5.4.1 Objective
5.4.2 Procedure
5.4.3 Summary for Tutorial 2
5.5 Summary
5.6 Additional Exercises
References
Appendix
Index