Multitemporal Earth Observation Image Analysis : Remote Sensing Image Sequences.

Format:

Book

Author/Creator:

Mallet, Clément.

Contributor:

Chehata, Nesrine.

Series:

ISTE Invoiced Series

Language:

English

Subjects (All):

Remote-sensing images.

Physical Description:

1 online resource (275 pages)

Edition:

1st ed.

Place of Publication:

Newark : John Wiley & Sons, Incorporated, 2024.

Summary:

Earth observation has witnessed a unique paradigm change in the last decade with a diverse and ever-growing number of data sources. Among them, time series of remote sensing images has proven to be invaluable for numerous environmental and climate studies. Multitemporal Earth Observation Image Analysis provides illustrations of recent methodological advances in data processing and information extraction from imagery, with an emphasis on the temporal dimension uncovered either by recent satellite constellations (in particular the Sentinels from the European Copernicus programme) or archival aerial images available in national archives. The book shows how complementary data sources can be efficiently used, how spatial and temporal information can be leveraged for biophysical parameter estimation, classification of land surfaces and object tracking, as well as how standard machine learning and state-of-the-art deep learning solutions can solve complex problems with real-world applications.

Contents:

Cover

Title Page

Copyright Page

Contents

Foreword

Chapter 1. Broader Application of the Time-SIFT Method: Proof-of-Concept of 3-D-Monitoring Study Cases with Various Spatiotemporal Scales

1.1. Introduction

1.2. The Time-SIFT method

1.2.1. General principle

1.2.2. Existing methods for multi-temporal photogrammetric processing

1.2.3. Step-by-step description of the Time-SIFT implementation

1.3. Case studies

1.3.1. La Peyne

1.3.2. Mount Rainier

1.3.3. Mornag's orchard

1.3.4. Soil bulk density

1.3.5. Benthic species in rocky shore

1.3.6. Test cases synthesis

1.4. Conclusion

1.5. References

Chapter 2. Hierarchical Crop Mapping from Satellite Image Sequences with Recurrent Neural Networks

2.1. Introduction

2.2. Literature

2.3. Background: sequence modeling with recurrent neural networks

2.3.1. Convolutional recurrent neural networks

2.3.2. Deep multi-layer RNNs

2.4. Hierarchical multi-stage convolutional recurrent network

2.4.1. Label refinement

2.4.2. Implementation details

2.5. Experiment

2.5.1. Dataset

2.5.2. Crop class hierarchy

2.5.3. Baselines and evaluation

2.5.4. Performance comparison

2.5.5. Ablation study

2.5.6. Simultaneous multi-level classification

2.5.7. Cloud robustness

2.6. Summary and future outlook

2.7. References

Chapter 3. Exploiting Multitemporal Multispectral High-resolution Satellite Data toward Annual Land Cover and Crop Type Mapping: A Case Study in Greece

3.1. Introduction

3.2. From raw data to analysis ready datasets

3.2.1. Availability of analysis ready data

3.2.2. Atmospheric correction

3.2.3. Directional effects and BRDF correction

3.2.4. Cloud and shadow detection

3.2.5. Geometric alignment and geolocation accuracy

3.2.6. Temporal gap-filling, composites and interpolated time series.

3.3. Classification and mapping

3.3.1. Reference data

3.3.2. Classification units and classifiers

3.3.3. Classification features

3.3.4. Accuracy assessment

3.3.5. Benchmarking classification features and stratification strategies

3.4. Data handling and computational challenges

3.5. Conclusions

3.6. Acknowledgments

3.7. References

Chapter 4. Irrigation Monitoring Using High Spatial and Temporal Resolutions Remote Sensing Time Series

4.1. Introduction

4.2. Fundamentals behind remote sensing for irrigation mapping

4.2.1. Optical remote sensing for irrigation mapping

4.2.2. SAR remote sensing for irrigation mapping

4.3. New methodologies for irrigation mapping using S1 and S2 time series

4.3.1. Decision tree algorithm for detecting irrigation events

4.3.2. Machine learning algorithms for mapping irrigated areas

4.3.3. Operational semi-supervised method for mapping irrigated areas

4.4. Limits and perspectives

4.4.1. Limits for irrigation mapping and irrigation detection using remote sensing time series

4.4.2. Perspectives in irrigation mapping using remote sensing time series data

4.5. Conclusions

4.6. References

Chapter 5. Trends in Satellite Time Series Processing for Vegetation Phenology Monitoring

5.1. Introduction

5.2. Time series processing for gap filling

5.2.1. Gap-filling methods

5.2.2. Assessment of gap-filling methods

5.3. Time series processing for phenology indicators estimation

5.4. Fusion of time series products for improved gap filling

5.4.1. MOGPR modeling

5.4.2. MOGPR assessment

5.4.3. MOGPR generalization

5.5. Time series processing toolbox: DATimeS

5.6. Discussion

5.7. Conclusions

5.8. Acknowledgments

5.9. References

Chapter 6. Data-Driven Spatio-Temporal Interpolation for Satellite-Derived Geophysical Tracers.

6.1. Notations

6.2. Introduction

6.3. Data assimilation

6.3.1. Introduction

6.3.2. Variational versus statistical DA

6.4. Data-driven methods

6.4.1. DINEOF schemes

6.4.2. Analog data assimilation

6.4.3. Neural network-based Kalman filters

6.4.4. Variational end-to-end learning-based methods

6.5. Application to satellite-derived ocean surface topography datasets

6.6. Conclusion and discussion

6.7. References

Chapter 7. Recent Advances in Tropical Cyclone Forecasting Using Machine Learning on Reanalysis and Remote Sensing

7.1. Background

7.1.1. Problem statement

7.1.2. Existing TC forecasts methods

7.1.3. Available data

7.1.4. Machine learning background

7.2. Handling spatiotemporal data for TC forecasting

7.2.1. Temporal feature structure: time series versus fixed time steps inputs

7.2.2. Spatial feature structure: fixed versus moving frame of reference

7.2.3. Combining different input data types

7.3. Application 1: intensity forecasting from spatiotemporal reanalysis, a hackathon experiment

7.3.1. Data and pipeline

7.3.2. Methods used by the participants

7.3.3. Results

7.4. Application 2: trajectory forecasting using fused deep learning

7.4.1. Data processing

7.4.2. Methodology: a Deep Fusion Model

7.4.3. Experimental evaluation

7.5. Applications using recurrent neural networks-convolutional neural networks

7.5.1. A hybrid CNN-LSTM model

7.5.2. An attention recurrent CNN model

7.6. Applications using remote sensing data

7.6.1. Deep learning intensity forecasting using microwave remote sensing

7.6.2. Deep learning intensity and wind structure forecasting using infrared remote sensing

7.7. Conclusion, current limitations and open problems

7.8. References

List of Authors

Index

EULA.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

9781394306657

1394306652

9781394306633

1394306636

OCLC:

1446125022