Change Detection and Image Time-Series Analysis 2
Supervised Methods
1st Edition
Published on 19. November 2021
272 pages
978-1-119-88227-5 (ISBN)
Description
Change Detection and Image Time Series Analysis 2 presents supervised machine-learning-based methods for temporal evolution analysis by using image time series associated with Earth observation data. Chapter 1 addresses the fusion of multisensor, multiresolution and multitemporal data. It proposes two supervised solutions that are based on a Markov random field: the first relies on a quad-tree and the second is specifically designed to deal with multimission, multifrequency and multiresolution time series.
Chapter 2 provides an overview of pixel based methods for time series classification, from the earliest shallow learning methods to the most recent deep-learning-based approaches.
Chapter 3 focuses on very high spatial resolution data time series and on the use of semantic information for modeling spatio-temporal evolution patterns.
Chapter 4 centers on the challenges of dense time series analysis, including pre processing aspects and a taxonomy of existing methodologies. Finally, since the evaluation of a learning system can be subject to multiple considerations,
Chapters 5 and 6 offer extensive evaluations of the methodologies and learning frameworks used to produce change maps, in the context of multiclass and/or multilabel change classification issues.
Chapter 2 provides an overview of pixel based methods for time series classification, from the earliest shallow learning methods to the most recent deep-learning-based approaches.
Chapter 3 focuses on very high spatial resolution data time series and on the use of semantic information for modeling spatio-temporal evolution patterns.
Chapter 4 centers on the challenges of dense time series analysis, including pre processing aspects and a taxonomy of existing methodologies. Finally, since the evaluation of a learning system can be subject to multiple considerations,
Chapters 5 and 6 offer extensive evaluations of the methodologies and learning frameworks used to produce change maps, in the context of multiclass and/or multilabel change classification issues.
More details
Other editions
Additional editions
Abdourrahmane M. Atto | Francesca Bovolo | Lorenzo Bruzzone
Change Detection and Image Time Series Analysis 2
Supervised Methods
Book
01/2022
1st Edition
ISTE Ltd
€164.50
Shipment within 3-4 weeks
Content
- Cover
- Half-Title Page
- Title Page
- Copyright Page
- Contents
- Preface
- List of Notations
- 1. Hierarchical Markov Random Fields for High Resolution Land Cover Classification of Multisensor and Multiresolution Image Time Series
- 1.1. Introduction
- 1.1.1. The role of multisensor data in time series classification
- 1.1.2. Multisensor and multiresolution classification
- 1.1.3. Previous work
- 1.2. Methodology
- 1.2.1. Overview of the proposed approaches
- 1.2.2. Hierarchical model associated with the first proposed method
- 1.2.3. Hierarchical model associated with the second proposed method
- 1.2.4. Multisensor hierarchical MPM inference
- 1.2.5. Probability density estimation through
- 1.3. Examples of experimental results
- 1.3.1. Results of the first method
- 1.3.2. Results of the second method
- 1.4. Conclusion
- 1.5. Acknowledgments
- 1.6. References
- 2. Pixel-based Classification Techniques for Satellite Image Time Series
- 2.1. Introduction
- 2.2. Basic concepts in supervised remote sensing classification
- 2.2.1. Preparing data before it is fed into classification algorithms
- 2.2.2. Key considerations when training supervised classifiers
- 2.2.3. Performance evaluation of supervised classifiers
- 2.3. Traditional classification algorithms
- 2.3.1. Support vector machines
- 2.3.2. Random forests
- 2.3.3. k-nearest neighbor
- 2.4. Classification strategies based on temporal feature representations
- 2.4.1. Phenology-based classification approaches
- 2.4.2. Dictionary-based classification approaches
- 2.4.3. Shapelet-based classification approaches
- 2.5. Deep learning approaches
- 2.5.1. Introduction to deep learning
- 2.5.2. Convolutional neural networks
- 2.5.3. Recurrent neural networks
- 2.6. References
- 3. Semantic Analysis of Satellite Image Time Series
- 3.1. Introduction
- 3.1.1. Typical SITS examples
- 3.1.2. Irregular acquisitions
- 3.1.3. The chapter structure
- 3.2. Why are semantics needed in SITS?
- 3.3. Similarity metrics
- 3.4. Feature methods
- 3.5. Classification methods
- 3.5.1. Active learning
- 3.5.2. Relevance feedback
- 3.5.3. Compression-based pattern recognition
- 3.5.4. Latent Dirichlet allocation
- 3.6. Conclusion
- 3.7. Acknowledgments
- 3.8. References
- 4. Optical Satellite Image Time Series Analysis for Environment Applications: From Classical Methods to Deep Learning and Beyond
- 4.1. Introduction
- 4.2. Annual time series
- 4.2.1. Overview of annual time series methods
- 4.2.2. Examples of annual times series analysis applications for environmental monitoring
- 4.2.3. Towards dense time series analysis
- 4.3. Dense time series analysis using all available data
- 4.3.1. Making dense time series consistent
- 4.3.2. Change detection methods
- 4.3.3. Summary and future developments
- 4.4. Deep learning-based time series analysis approaches
- 4.4.1. Recurrent Neural Network (RNN) for Satellite Image Time Series
- 4.4.2. Convolutional Neural Networks (CNN) for Satellite Image Time Series
- 4.4.3. Hybrid models: Convolutional Recurrent Neural Network (ConvRNN) models for Satellite Image Time Series
- 4.4.4. Synthesis and future developments
- 4.5. Beyond satellite image time series and deep learning: convergence between time series and video approaches
- 4.5.1. Increased image acquisition frequency: from time series to spaceborne time-lapse and videos
- 4.5.2. Deep learning and computer vision as technology enablers
- 4.5.3. Future steps
- 4.6. References
- 5. A Review on Multi-temporal Earthquake Damage Assessment Using Satellite Images
- 5.1. Introduction
- 5.1.1. Research methodology and statistics
- 5.2. Satellite-based earthquake damage assessment
- 5.3. Pre-processing of satellite images before damage assessment
- 5.4. Multi-source image analysis
- 5.5. Contextual feature mining for damage assessment
- 5.5.1. Textural features
- 5.5.2. Filter-based methods
- 5.6. Multi-temporal image analysis for damage assessment
- 5.6.1. Use of machine learning in damage assessment problem
- 5.6.2. Rapid earthquake damage assessment
- 5.7. Understandingdamage followingan earthquakeusing satellite-based SAR
- 5.7.1. SAR fundamental parameters and acquisition vector
- 5.7.2. Coherent methods for damage assessment
- 5.7.3. Incoherent methods for damage assessment
- 5.7.4. Post-earthquake-only SAR data-based damage assessment
- 5.7.5. Combination of coherent and incoherent methods for damage assessment
- 5.7.6. Summary
- 5.8. Use of auxiliary data sources
- 5.9. Damage grades
- 5.10. Conclusion and discussion
- 5.11. References
- 6. Multiclass Multilabel Change of State Transfer Learning from Image Time Series
- 6.1. Introduction
- 6.2. Coarse- to fine-grained change of state dataset
- 6.3. Deep transfer learning models for change of state classification
- 6.3.1. Deep learning model library
- 6.3.2. Graph structures for the CNN library
- 6.3.3. Dimensionalities of the learnables for the CNN library
- 6.4. Change of state analysis
- 6.4.1. Transfer learning adaptations for the change of state classification issues
- 6.4.2. Experimental results
- 6.5. Conclusion
- 6.6. Acknowledgments
- 6.7. References
- List of Authors
- Index
- Summary of Volume 1
- EULA
