Introductory Digital Image Processing

Series
Prentice Hall
Author
John R Jensen  
Publisher
Pearson
Cover
Softcover
Edition
4
Language
English
Total pages
544
Pub.-date
April 2015
ISBN13
9780134058160
ISBN
013405816X
Related Titles



Description

For junior/graduate-level courses in Remote Sensing in Geography, Geology, Forestry, and Biology.

Introductory Digital Image Processing: A Remote Sensing Perspective focuses on digital image processing of aircraft- and satellite-derived, remotely sensed data for Earth resource management applications. Extensively illustrated, it explains how to extract biophysical information from remote sensor data for almost all multidisciplinary land-based environmental projects. Part of the Pearson Series Geographic Information Science.


Now in full color, the Fourth Edition provides up-to-date information on analytical methods used to analyze digital remote sensing data. Each chapter contains a substantive reference list that can be used by students and scientists as a starting place for their digital image processing project or research. A new appendix provides sources of imagery and other geospatial information.

 

Features

  • Now presented in full color. Contains new, specially developed detailed graphics consistent with current technology.
  • Emphasizes the application of digital image processing algorithms rather than engineering "signal processing"—Extracts useful Earth resource information from remotely sensed imagery.
  • Organizes content according to the general flow or method by which digital remote sensor data is actually analyzed.
  • Presents all algorithms in relatively simple algebra terms.
  • Summarizes the "remote sensing process"—Includes hypothesis testing procedures, data collection, data analysis, and information presentation (display) alternatives.
  • Reviews new methods of image digitization and data compression and summarizes new formats for digital imagery—Includes the national aerial photography program (NAPP); multispectral imaging using discrete detectors and scanning mirrors; multispectral imaging using linear arrays; imagine spectrometry using linear and area arrays; the proposed Earth Observing System (EOS).
  • Provides a substantive reference list in each chapter—Enables readers to obtain additional information on a topic.
  • Surveys the state-of-the- art of digital image processing hardware and software configurations—Using mainframe, workstation, and personal computers.
  • Discusses the functionality of 25 commercial and 6 public digital image processing systems.
  • Contains a more detailed treatment of univariate and multivariate statistics which are routinely extracted from remotely sensed data.
  • Introduces the concept of "scientific visualization"and presents alternatives for producing scaled, color, hardcopy output.
  • Contains detailed information on how to radiometrically correct for atmospheric attenuation in remotely sensed data using "relative image normalization" and "absolute radiometric correction" techniques.
  • Describes how linear and non-linear contrast enhancement is performed—Provides in-depth coverage of histogram equalization.
  • Contains a section on spatial filtering in the frequency domain using the Fourier Transform.
  • Introduces classification schemes, including the new NOAA CoastWatch scheme.
  • Describes all the major supervised classification algorithms and methods for making them efficient—Also shows how to incorporate a priori and a posteriori probabilities.
  • Features a section based on the ISODATA clustering algorithm—Includes informative graphics which describe how the feature space is iteratively partitioned.
  • Contains a flow diagram of the general steps required to perform digital change detection of remotely sensed data—Presents 9 change detection algorithms along with diagrams which depict exactly how to perform the change direction.
  • Includes a complete description of the major vector (e.g., TIGER, DLG) and raster data sets (e.g., remotely sensed data) available as well as a discussion of the various data analysis functions typically available in a GIS.
  • Includes a case study which predicts the geographic distribution of wetlands in a freshwater lake based on the GIS analysis of various biophysical variables.
  • NEW! A new appendix is provided that contains a list of selected geospatial datasets that can be evaluated and/or downloaded via the Internet, including: digital elevation information, hydrology, land use/land cover and biodiversity/habitat, road network and population demographic data, and several types of publicly- and commercially-available remote sensor-data. Map or image examples of the datasets are presented where appropriate.

Keep your course current and relevant

Content updates

Chapter 1: Introduction

Greater emphasis is now placed on the importance of ground reference information that can be used to calibrate remote sensor data and assess the accuracy of remote sensing-derived products such as thematic maps. The “Remote Sensing Process” has been updated to reflect recent innovations in digital image processing. Greater emphasis is now placed on the use of remote sensing to solve local, high-spatial resolution problems as well as for use in global climate change research. This chapter now includes detailed information about the increasing demand for people trained in remote sensing digital image processing. Information is provided from a) the NRC (2013) Future U.S. Workforce for Geospatial Intelligence study, and b) U.S. Department of Labor Employment and Training Administration (USDOLETA, 2014) data about the 39,900 “Remote Sensing Scientists and Technologists” and “Remote Sensing Technicians” job openings projected from 2012–2022. Most of these occupations require training in remote sensing digital image processing.

 

Chapter 2: Remote Sensing Data

Collection

This chapter provides information about historical, current, and projected sources of remotely sensed data.

Detailed information about new and proposed satellite remote sensing systems (e.g., Astrium’s Pleiades and

SPOT 6; DigitalGlobe’s GeoEye-1, GeoEye-2, World-View-1, WorldView-2, WorldView-3; India’s CartoSat

and ResourceSat; Israel’s EROS A2; Korea’s KOMPSAT; NASA’s Landsat 8; NOAA’s NPOESS;

RapidEye, etc.) and airborne remote sensing systems (e.g., PICTOMETRY, Microsoft’s UltraCAM, Leica’s

Airborne Digital System 80) are included in the fourth edition. Technical details about decommissioned (e.g.,

SPOT 1, 2; Landsat 5), degraded (e.g., Landsat 7 ETM+) or failed (e.g., European Space Agency Envisat)

sensor systems are provided.

 

Chapter 3: Digital Image

Processing Hardware and

Software

As expected, the computer hardware (e.g., CPUs, RAM, mass storage, digitization technology, displays,

transfer/storage technology) and software [e.g., multispectral, hyperspectral, per-pixel, object-based image

analysis (OBIA)] necessary to perform digital image processing have progressed significantly since the last

edition. Improvements in computer hardware often used to perform digital image processing are discussed.

The most important functions, characteristics and sources of the major digital image processing software

are provided.

 

Chapter 4: Image Quality

Basic digital image processing mathematical notation is reviewed along with the significance of the histogram.

The importance of metadata is introduced. Visual methods of assessing image quality are presented including three-dimensional representation. Univariate and multivariate methods of assessing the initial quality of digital remote sensor data are refreshed. A new section on geostatistical analysis, autocorrelation and kriging interpolation is provided.

 

Chapter 5: Display Alternatives

and Scientific Visualization

New information is provided on: liquid crystal displays (LCD), image compression alternatives, color coordinate

systems (RGB, Intensity-Hue-Saturation, and Chromaticity), the use of 8- and 24-bit color look-up tables, and new methods of merging (fusing) different types of imagery (e.g., Gram-Schmidt, regression Kriging). Additional information is provided about measuring distance, perimeter, shape and polygon area using digital imagery.

 

Chapter 6: Radiometric Correction

Additional information is provided about electromagnetic radiation principles (e.g., Fraunhofer absorption

features) and the spectral reflectance characteristics of selected natural and human-made materials. Updated

information about the most important radiometric correction algorithms is provided, including: a) those

that perform absolute radiometric correction (e.g., MODTRAN 4, ACORN, FLAASH, QUAC, ATCOR,

empirical line calibration) and, b) those that perform relative radiometric correction (e.g., single and multiple-

date image normalization).

 

Chapter 7: Geometric Correction

Traditional as well as improved methods of image-to map rectification and image-to-image registration are

provided. In addition, this edition contains an expanded discussion on developable surfaces and the properties

and advantages/disadvantages of several of the most heavily used cylindrical, azimuthal, and conical map projections. MODIS satellite imagery is projected using selected map projections (e.g., Mercator, Lambert

Azimuthal Equal-area). The image mosaicking section contains new examples and demonstrates the characteristics of the USGS annual mosaic of Landsat ETM+ data (i.e., the WELD: Web-enabled Landsat Data project).

 

Chapter 8: Image Enhancement

The image magnification and reduction sections are revised. In addition, the following image enhancement

techniques are updated: band ratioing, neighborhood raster operations, spatial convolution filtering and

edge enhancement, frequency filtering, texture extraction, and Principal Components Analysis (PCA). The

vegetation indices (VI) section has been significantly revised to include new information on the dominant

factors controlling leaf reflectance and the introduction of numerous new indices with graphic examples.

Several new texture transforms are introduced (e.g., Moran’s I Spatial Autocorrelation) and new information

is provided on the extraction of texture from images using Grey-level Co-occurrence Matrices (GLCM).

The chapter concludes with a new discussion on landscape ecology metrics that can be extracted from remotely

sensed data.

 

Chapter 9: Information Extraction Using Pattern Recognition

Updated information on the American Planning Association Land-Based Classification Standard (NLCS),

the U.S. National Land Cover Database (NLCD) Classification System, NOAA’s Coastal Change Analysis

Program (C-CAP) Classification Scheme, and the IGBP Land-Cover Classification System is included.

New methods of feature (band) selection are introduced (e.g., Correlation Matrix Feature Selection). Additional

information is provided on Object-based Image Analysis (OBIA) classification methods, including

new OBIA application examples.

 

Chapter 10: Information Extraction Using Artificial Intelligence

New information is provided on image classification using machine-learning decision trees, regression trees,

Random Forest (trees), and Support Vector Machines (SVM). Detailed information is now provided on a

number of machine-learning, data-mining decision tree/regression tree programs that can be used to develop

production rules (e.g., CART, S-Plus, R Development Core Team, C4.5, C5.0, Cubist). New information about advances in neural network analysis of remote sensor data is included for Multi-layer Perceptrons, Kohonen’s Self-Organizing Map, and fuzzy ARTMAP neural networks. A new discussion about the advantages and disadvantages of artificial neural networks is provided.

 

Chapter 11: Information Extraction Using Imaging Spectroscopy

Advances in airborne and satellite hyperspectral data collection are discussed. Advances in the methods used

to process and analyze hyperspectral imagery are provided, including: end-member selection and analysis, mapping algorithms, Spectral Mixture Analysis (SMA), continuum removal, spectroscopic library matching techniques, machine-learning hyperspectral analysis techniques, new hyperspectral indices, and derivative spectroscopy.

 

Chapter 12: Change Detection

This book has always contained detailed digital change detection information. New information is provided on

the impact of sensor system look angle and amount of tree or building obscuration. Advances in binary “change/no-change” algorithms are provided including new analytical methods used to identify the change thresholds and new commercial change detection products such as ESRI’s Change Matters and MDA’s National Urban Change Indicator. Significant advances in thematic “from-to” change detection algorithms are discussed including photogrammetric and LiDARgrammetric change detection, OBIA post-classification comparison change detection, and Neighborhood Correlation Image (NCI) change detection.

 

Chapter 13: Remote Sensing derived Thematic Map Accuracy Assessment

There is a significant amount of literature and debate about the best method(s) to use to determine the accuracy

of remote sensing-derived thematic map produced from a single date of imagery or a thematic map derived

from multiple dates of imagery (i.e., change detection). The accuracy assessment alternatives and

characteristics of the debate are discussed more thoroughly.

 

Appendix: Sources of Imagery and other Geospatial Information

A new appendix is provided that contains a list of selected geospatial datasets that can be evaluated and/

or downloaded via the Internet, including: digital elevation information, hydrology, land use/land cover and

biodiversity/habitat, road network and population demographic data, and several types of publicly- and

commercially-available remote sensor-data. Map or image examples of the datasets are presented where appropriate.

 

New to this Edition

Keep your course current and relevant

Content updates

Chapter 1: Introduction

Greater emphasis is now placed on the importance of ground reference information that can be used to calibrate remote sensor data and assess the accuracy of remote sensing-derived products such as thematic maps. The “Remote Sensing Process” has been updated to reflect recent innovations in digital image processing. Greater emphasis is now placed on the use of remote sensing to solve local, high-spatial resolution problems as well as for use in global climate change research. This chapter now includes detailed information about the increasing demand for people trained in remote sensing digital image processing. Information is provided from a) the NRC (2013) Future U.S. Workforce for Geospatial Intelligence study, and b) U.S. Department of Labor Employment and Training Administration (USDOLETA, 2014) data about the 39,900 “Remote Sensing Scientists and Technologists” and “Remote Sensing Technicians” job openings projected from 2012–2022. Most of these occupations require training in remote sensing digital image processing.

 

Chapter 2: Remote Sensing Data

Collection

This chapter provides information about historical, current, and projected sources of remotely sensed data.

Detailed information about new and proposed satellite remote sensing systems (e.g., Astrium’s Pleiades and

SPOT 6; DigitalGlobe’s GeoEye-1, GeoEye-2, World-View-1, WorldView-2, WorldView-3; India’s CartoSat

and ResourceSat; Israel’s EROS A2; Korea’s KOMPSAT; NASA’s Landsat 8; NOAA’s NPOESS;

RapidEye, etc.) and airborne remote sensing systems (e.g., PICTOMETRY, Microsoft’s UltraCAM, Leica’s

Airborne Digital System 80) are included in the fourth edition. Technical details about decommissioned (e.g.,

SPOT 1, 2; Landsat 5), degraded (e.g., Landsat 7 ETM+) or failed (e.g., European Space Agency Envisat)

sensor systems are provided.

 

Chapter 3: Digital Image

Processing Hardware and

Software

As expected, the computer hardware (e.g., CPUs, RAM, mass storage, digitization technology, displays,

transfer/storage technology) and software [e.g., multispectral, hyperspectral, per-pixel, object-based image

analysis (OBIA)] necessary to perform digital image processing have progressed significantly since the last

edition. Improvements in computer hardware often used to perform digital image processing are discussed.

The most important functions, characteristics and sources of the major digital image processing software

are provided.

 

Chapter 4: Image Quality

Basic digital image processing mathematical notation is reviewed along with the significance of the histogram.

The importance of metadata is introduced. Visual methods of assessing image quality are presented including three-dimensional representation. Univariate and multivariate methods of assessing the initial quality of digital remote sensor data are refreshed. A new section on geostatistical analysis, autocorrelation and kriging interpolation is provided.

 

Chapter 5: Display Alternatives

and Scientific Visualization

New information is provided on: liquid crystal displays (LCD), image compression alternatives, color coordinate

systems (RGB, Intensity-Hue-Saturation, and Chromaticity), the use of 8- and 24-bit color look-up tables, and new methods of merging (fusing) different types of imagery (e.g., Gram-Schmidt, regression Kriging). Additional information is provided about measuring distance, perimeter, shape and polygon area using digital imagery.

 

Chapter 6: Radiometric Correction

Additional information is provided about electromagnetic radiation principles (e.g., Fraunhofer absorption

features) and the spectral reflectance characteristics of selected natural and human-made materials. Updated

information about the most important radiometric correction algorithms is provided, including: a) those

that perform absolute radiometric correction (e.g., MODTRAN 4, ACORN, FLAASH, QUAC, ATCOR,

empirical line calibration) and, b) those that perform relative radiometric correction (e.g., single and multiple-

date image normalization).

 

Chapter 7: Geometric Correction

Traditional as well as improved methods of image-to map rectification and image-to-image registration are

provided. In addition, this edition contains an expanded discussion on developable surfaces and the properties

and advantages/disadvantages of several of the most heavily used cylindrical, azimuthal, and conical map projections. MODIS satellite imagery is projected using selected map projections (e.g., Mercator, Lambert

Azimuthal Equal-area). The image mosaicking section contains new examples and demonstrates the characteristics of the USGS annual mosaic of Landsat ETM+ data (i.e., the WELD: Web-enabled Landsat Data project).

 

Chapter 8: Image Enhancement

The image magnification and reduction sections are revised. In addition, the following image enhancement

techniques are updated: band ratioing, neighborhood raster operations, spatial convolution filtering and

edge enhancement, frequency filtering, texture extraction, and Principal Components Analysis (PCA). The

vegetation indices (VI) section has been significantly revised to include new information on the dominant

factors controlling leaf reflectance and the introduction of numerous new indices with graphic examples.

Several new texture transforms are introduced (e.g., Moran’s I Spatial Autocorrelation) and new information

is provided on the extraction of texture from images using Grey-level Co-occurrence Matrices (GLCM).

The chapter concludes with a new discussion on landscape ecology metrics that can be extracted from remotely

sensed data.

 

Chapter 9: Information Extraction Using Pattern Recognition

Updated information on the American Planning Association Land-Based Classification Standard (NLCS),

the U.S. National Land Cover Database (NLCD) Classification System, NOAA’s Coastal Change Analysis

Program (C-CAP) Classification Scheme, and the IGBP Land-Cover Classification System is included.

New methods of feature (band) selection are introduced (e.g., Correlation Matrix Feature Selection). Additional

information is provided on Object-based Image Analysis (OBIA) classification methods, including

new OBIA application examples.

 

Chapter 10: Information Extraction Using Artificial Intelligence

New information is provided on image classification using machine-learning decision trees, regression trees,

Random Forest (trees), and Support Vector Machines (SVM). Detailed information is now provided on a

number of machine-learning, data-mining decision tree/regression tree programs that can be used to develop

production rules (e.g., CART, S-Plus, R Development Core Team, C4.5, C5.0, Cubist). New information about advances in neural network analysis of remote sensor data is included for Multi-layer Perceptrons, Kohonen’s Self-Organizing Map, and fuzzy ARTMAP neural networks. A new discussion about the advantages and disadvantages of artificial neural networks is provided.

 

Chapter 11: Information Extraction Using Imaging Spectroscopy

Advances in airborne and satellite hyperspectral data collection are discussed. Advances in the methods used

to process and analyze hyperspectral imagery are provided, including: end-member selection and analysis, mapping algorithms, Spectral Mixture Analysis (SMA), continuum removal, spectroscopic library matching techniques, machine-learning hyperspectral analysis techniques, new hyperspectral indices, and derivative spectroscopy.

 

Chapter 12: Change Detection

This book has always contained detailed digital change detection information. New information is provided on

the impact of sensor system look angle and amount of tree or building obscuration. Advances in binary “change/no-change” algorithms are provided including new analytical methods used to identify the change thresholds and new commercial change detection products such as ESRI’s Change Matters and MDA’s National Urban Change Indicator. Significant advances in thematic “from-to” change detection algorithms are discussed including photogrammetric and LiDARgrammetric change detection, OBIA post-classification comparison change detection, and Neighborhood Correlation Image (NCI) change detection.

 

Chapter 13: Remote Sensing derived Thematic Map Accuracy Assessment

There is a significant amount of literature and debate about the best method(s) to use to determine the accuracy

of remote sensing-derived thematic map produced from a single date of imagery or a thematic map derived

from multiple dates of imagery (i.e., change detection). The accuracy assessment alternatives and

characteristics of the debate are discussed more thoroughly.

 

Appendix: Sources of Imagery and other Geospatial Information

A new appendix is provided that contains a list of selected geospatial datasets that can be evaluated and/

or downloaded via the Internet, including: digital elevation information, hydrology, land use/land cover and

biodiversity/habitat, road network and population demographic data, and several types of publicly- and

commercially-available remote sensor-data. Map or image examples of the datasets are presented where appropriate.

Table of Contents

1. Remote Sensing and Digital Image Processing
2. Remote Sensing Data Collection
3. Digital Image Processing Hardware and Software
4. Image Quality Assessment and Statistical Evaluation  
5. Display Alternatives and Scientific Visualization
6. Electromagnetic Radiation Principles and Radiometric  Correction 
7. Geometric Correction
8. Image Enhancement  
9. Thematic Information Extraction: Pattern Recognition
10. Information Extraction Using Artificial Intelligence
11. Information Extraction Using Imaging Spectroscopy
12. Change Detection
13. Remote Sensing–Derived Thematic Map Accuracy Assessment

Appendix: Sources of Imagery and Other Geospatial Information

Index

 

 

Author

John R. Jensen received a BA in geography from California State University at Fullerton, an MS from Brigham Young University (BYU), and a PhD from the University of California at Los Angeles (UCLA). He is a Carolina Distinguished Professor Emeritus in the Department of Geography at the University of SouthCarolina. He is a certified photogrammetrist and a past president of the American Society for Photogrammetry & Remote Sensing (ASP&RS): The Geospatial Information Society. He has conducted more than 50 remote sensing-related projects sponsored by NASA, DOE, NOAA, and the Nature Conservancy and published more than 120 refereed journal articles. He mentored 34 PhD and 62 master’s students. He received the SAIC/ASP&RS John E. Estes Memorial Teaching Award for education, mentoring, and training in remote sensing and GIS. He received the U.S. Geological Survey/National Aeronautics & Space Administration (NASA) William T. Pecora Award for his remote sensing research contributions. He received the Association of American Geographers (AAG) Lifetime Achievement Award for research and education in remote sensing and GIScience. He was the Editor-in-chief of the journal GIScience & Remote Sensing published by Taylor & Francis. He is co-author of Introductory Geographic Information Systems and author of Remote Sensing of the Environment: An Earth Resource Perspective, 2nd edition, also published by Pearson. He has been associated with eight National Research Council (NRC) remote sensing-related committees and subsequent National Academy Press publications. He became an Honorary Member of ASP&RS in 2013, the highest award bestowed by ASP&RS.