Image Analysis, Classification and Change Detection in Remote Sensing

With Algorithms for Python, Fourth Edition

Morton John Canty

• Publisher: CRC Press
• Year: 2019
• Print ISBN: 9781138613225
• eBook ISBN: 9780429875342
• Edition: 4
• Topic: Physical Sciences
• Subtopic: Arithmetic

Image Analysis, Classification and Change Detection in Remote Sensing: With Algorithms for Python, Fourth Edition, is focused on the development and implementation of statistically motivated, data-driven techniques for digital image analysis of remotely sensed imagery and it features a tight interweaving of statistical and machine learning theory of algorithms with computer codes. It develops statistical methods for the analysis of optical/infrared and synthetic aperture radar (SAR) imagery, including wavelet transformations, kernel methods for nonlinear classification, as well as an introduction to deep learning in the context of feed forward neural networks.

New in the Fourth Edition:

• 
  
  

• An in-depth treatment of a recent sequential change detection algorithm for polarimetric SAR image time series.

• 
  
  

• The accompanying software consists of Python (open source) versions of all of the main image analysis algorithms.

• 
  
  

• Presents easy, platform-independent software installation methods (Docker containerization).

• 
  
  

• Utilizes freely accessible imagery via the Google Earth Engine and provides many examples of cloud programming (Google Earth Engine API).

• 
  
  

• Examines deep learning examples including TensorFlow and a sound introduction to neural networks,

Based on the success and the reputation of the previous editions and compared to other textbooks in the market, Professor Canty's fourth edition differs in the depth and sophistication of the material treated as well as in its consistent use of computer codes to illustrate the methods and algorithms discussed. It is self-contained and illustrated with many programming examples, all of which can be conveniently run in a web browser. Each chapter concludes with exercises complementing or extending the material in the text.

1

Images, Arrays, and Matrices

There are many Earth observation satellite-based sensors, both active and passive, currently in orbit or planned for the near future. Representative of these, we describe briefly the multispectral ASTER system (Abrams et al., 1999) and the TerraSAR-X synthetic aperture radar satellite (Pitz and Miller, 2010). See Jensen (2018), Richards (2012) and Mather and Koch (2010) for overviews of remote sensing satellite platforms.

The Advanced Spaceborne Thermal Emission and Reflectance Radiometer (ASTER) instrument was launched in December 1999 on the Terra spacecraft. It is being used to obtain detailed maps of land surface temperature, reflectance and elevation and consists of sensors to measure reflected solar radiance and thermal emission in three spectral intervals:

• VNIR: Visible and near-infrared bands 1, 2, 3N, and 3B, in the spectral region between 0.52 and 0.86 μm (four arrays of charge-coupled detectors (CCDs) in pushbroom scanning mode).

• SWIR: Short wavelength infrared bands 4 to 9 in the region between 1.60 and 2.43 μm (six cooled PtSi-Si Schottky barrier arrays, pushbroom scanning).

• TIR: Thermal infrared bands 10 to 14 covering a spectral range from 8.13 to 11.65 μm (cooled HgCdTe detector arrays, whiskbroom scanning).

The altitude of the spacecraft is 705 km. The across- and in-track ground sample distances (GSDs), i.e., the detector widths projected through the system optics onto the Earth’s surface, are 15 m (VNIR), 30 m (SWIR) and 90 m (TIR).* The telescope associated with the 3B sensors is back-looking at an angle of 27.6° to provide, together with the 3N sensors, along-...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface to the First Edition
• Preface to the Second Edition
• Preface to the Third Edition
• Preface to the Fourth Edition
• Author Biography
• 1 Images, Arrays, and Matrices
• 2 Image Statistics
• 3 Transformations
• 4 Filters, Kernels, and Fields
• 5 Image Enhancement and Correction
• 6 Supervised Classification Part 1
• 7 Supervised Classification Part 2
• 8 Unsupervised Classification
• 9 Change Detection
• A Mathematical Tools
• B Efficient Neural Network Training Algorithms
• C Software
• Mathematical Notation
• References
• Index