Assessing the accuracy of remotely sensed data : principles and practices

Cover
Half Title
Title Page
Copyright Page
Dedication Page
Contents
Preface
Acknowledgments
Authors
Chapter 1: Introduction
	Why Map?
	Why Assess the Accuracy of a Map?
	Types of Map Accuracy Assessment
	Critical Steps in Accuracy Assessment
	Organization of the Book
Chapter 2: The History of Map Accuracy Assessment
	How Maps Are Made
	History of Accuracy Assessment
	Positional Accuracy Assessment
	Thematic Accuracy Assessment
Chapter 3: Planning for Conducting an Accuracy Assessment
	What Type of Assessment?
		Qualitative Accuracy Assessment
		Between Qualitative and Quantitative Assessment
			Similarity Analysis
			Error Budget
		Quantitative Accuracy Assessment
			Positional Accuracy Assessment
			Thematic Accuracy Assessment
	Why Do the Assessment?
	How Will the Assessment Be Used?
	What Special Considerations Exist?
	Conclusions
Chapter 4: Positional Accuracy
	Introduction
	What Is Positional Accuracy?
	What are the Common Standards for Positional Accuracy?
	How Should Positional Accuracy Assessment Be Designed and the Samples Be Selected?
	How Is Positional Accuracy Analyzed?
	Review of Basic Statistics
		Parameters and Statistics
	Estimating the Dispersal of Variables
	Estimating a Confidence Interval around the Estimate of the Mean
	Statistics in Positional Accuracy Assessment
		If the Errors Are Assumed to Be Normally Distributed
		If the Errors Are Assumed Not to Be Normally Distributed
	How Is Positional Accuracy Reported?
	Summary
Chapter 5: Thematic Map Accuracy Basics
	Non-site-specific Assessments
	Site-specific Assessments
		The Error Matrix
			Overall, Producer’ s, and User’ s Accuracies
			Mathematical Representation of the Error Matrix
Chapter 6: Thematic Map Accuracy Assessment Considerations
	What Are the Thematic Map Classes to Be Assessed?
		The Classification Scheme
		Other Data Considerations
			Continuous versus Non-Continuous Data
			Spatial Autocorrelation
	What Is the Appropriate Sample Unit?
		Single Pixel
		Cluster of Pixels as a Single Sample Unit
		Polygon as Single Sample Unit
		Clusters of Sample Units
	How Many Samples Should Be Taken?
		Binomial Distribution
		Multinomial Distribution
	How Should the Samples Be Chosen?
		Sampling Schemes
		Sampling Scheme Considerations
			Equal versus Minimum versus Proportional Distribution
	Conclusions
Chapter 7: Reference Data Collection
	What Should Be the Source of the Reference Data?
		Using Existing versus Newly Collected Data
		Remotely Sensed Data versus Field Visits
	How Should the Reference Data Be Collected?
	When Should the Reference Data Be Collected?
	Ensuring Objectivity and Consistency
		Data Independence
		Data Collection Consistency
		Quality Control
Chapter 8: Basic Analysis Techniques
	Kappa
	Margfit
	Conditional Kappa
	Weighted Kappa
	Compensation for Chance Agreement
	Confidence Limits
	Area Estimation/Correction
Chapter 9: Analysis of Differences in the Error Matrix
	Errors in the Reference Data
	Sensitivity of the Classification Scheme to Observer Variability
	Inappropriateness of the Remote Sensing Data Employed to Make the Map
	Mapping Error
	Summary
Chapter 10: Fuzzy Accuracy Assessment
	Expanding the Major Diagonal of the Error Matrix
	Measuring Map Class Variability
	The Fuzzy Error Matrix Approach
		The Fuzzy Error Matrix
		Implementation of the Fuzzy Error Matrix
		Another Fuzzy Error Matrix Example
	Summary
Chapter 11: Object-Based or Polygon Accuracy Assessment
	What Sample Unit Should Be Used?
	Traditional Tally-based versus Area-based Error Matrix
		Comparison of the Two Error Matrix Approaches
	Reference Data Collection/Labeling
	Computing the Accuracy
	Conclusions
Chapter 12: An Object-based Accuracy Assessment Case Study: The Grand Canyon National Park/Grand Canyon-Parashant National Monument Vegetation Classification and Mapping Project
	Introduction
	Overview of the Case Study
	Accuracy Assessment
		What Are the Thematic Classes to Be Assessed?
		What Is the Appropriate Sampling Unit?
		How Many Samples Should Be Taken?
		How Should the Samples Be Chosen?
		What Should Be the Source of the Reference Data?
		How Should the Reference Data Be Collected?
		When Should the Reference Data Be Collected?
		How Do I Ensure Consistency and Objectivity in My Data Collection?
	Analysis
		What Is an Error Matrix and How Should It Be Used?
		What Are the Statistical Properties Associated with the Error Matrix and What Analysis Techniques Are Applicable?
		What Is Fuzzy Accuracy and How Can You Conduct a Fuzzy Accuracy Assessment?
	Results
	Lessons Learned
Chapter 13: The California Hardwood Rangeland Project
	Introduction
	Background
	Sample Design
	How Is the Map Information Distributed?
	What Is the Appropriate Sample Unit?
		How Many Samples Should Be Taken?
		How Should the Samples Be Chosen and Distributed across the Landscape?
		Samples Chosen from the 1981 Map
		Samples Chosen from the 1990 Coverage
	Reference Data Collection
		What Should Be the Source Data for the Reference Samples?
		What Type of Information Should Be Collected?
		When Should the Reference Data Be Collected?
			Quality Control
	Analysis
		Development of the Error Matrices
		Statistical Analysis
		Analysis of Off-diagonal Samples
		Crown Closure Analysis
		Crown Closure Map Results
		Cover Type Analysis
		Cover Type Map Results
		Extent
	Discussion
	Conclusions
Chapter 14: Advanced Topics
	Change Detection
		Reference Data
		Sampling
		Change Detection Error Matrix
		Two-Step Approach to Change Detection Accuracy Assessment
		Case Study
			Step 1: Accuracy of the Change Areas
			Step 2: Change/No Change Assessment
	Multi-layer Assessments
Chapter 15: Summary and Conclusions
Appendix 1: ASPRS Positional Accuracy Standards for Digital Geospatial Data
Foreword
ASPRS Positional Accuracy Standards for Digital Geospatial Data
Annex A: Background and Justifications (Informative)
Annex B: Data Accuracy and Quality Examples (Normative)
Annex C: Accuracy Testing and Reporting Guidelines (Normative)
Annex D: Accuracy Statistics and Example (Normative)
Bibliography
Index