Geospatial Intelligence: Applications and Future Trends

✍ Scribed by Fatimazahra Barramou, El Hassan El Brirchi, Khalifa Mansouri, Youness Dehbi

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 180
• Series: Advances in Science, Technology & Innovation
• Edition: 1
• Category: Library

✦ Synopsis

This book explores cutting-edge methods combining geospatial technologies and artificial intelligence related to several fields such as smart farming, urban planning, geology, transportation, and 3D city models. It introduces techniques which range from machine and deep learning to remote sensing for geospatial data analysis.

The book consists of two main parts that include 13 chapters contributed by promising authors. The first part deals with the use of artificial intelligence techniques to improve spatial data analysis, whereas the second part focuses on the use of artificial intelligence with remote sensing in various fields. Throughout the chapters, the interest for the use of artificial intelligence is demonstrated for different geospatial technologies such as aerial imagery, drones, Lidar, satellite remote sensing, and more.

The work in this book is dedicated to the scientific community interested in the coupling of geospatial technologies and artificial intelligence and exploring the synergetic effects of both fields. It offers practitioners and researchers from academia, the industry and government information, experiences and research results about all aspects of specialized and interdisciplinary fields on geospatial intelligence.

✦ Table of Contents

Preface
Contents
Contributors
Spatial Data and Artificiel Intelligence
1 Towards a Multi-agents Model for Automatic Big Data Processing to Support Urban Planning
Abstract
1 Introduction
2 Background of the Study
2.1 Big Data
2.2 Geospatial Big Data
2.3 Smart Data
2.4 Machine Learning
2.4.1 Supervised Machine Learning
2.4.2 Semi-supervised Machine Learning
2.4.3 Applications of Machine Learning
2.4.4 Metrics
2.5 Multi-agent System
3 Related Works
3.1 Big Data Analytics Processes
3.2 Processes Automation Through Automatic Service Composition ASC
3.3 Discussion
4 Proposed Approach
4.1 Overview of the Functionalities of the Proposed Multi-agent System
4.2 The Proposed Multi-agent System Workflow
5 Tests and Results
5.1 Data Set: Individuals, Households
5.2 Results and Metrics
5.3 Data Visualization
6 Conclusion
References
2 Geospatial Forecasting and Social Media Exploration Based on Sentiment Analysis: Application to Flood Forecasting
Abstract
1 Introduction
2 State of the Art
2.1 Geospatial Aspect
2.2 Geospatial Forecasting and Social Media
2.3 Sentiment Analysis for Social Media and Geospatial Research
3 Our Approach
3.1 Twitter Search Filtering
3.2 Tweet Analysis
3.3 Geographic-Temporal Predictive Method
3.4 Results and Validation
4 Conclusion
References
3 Deep Convolution Neural Network for Automated Method of Road Extraction on Aerial Imagery
Abstract
1 Introduction
2 Related Work
2.1 Semantic Segmentation
2.2 Deep Learning for Remote Sensing
2.3 Deep Learning for Road Detection
3 Methodology
3.1 Dataset Description
3.2 Data Augmentation
3.3 Network Architecture
4 Experiments and Results
4.1 Metrics for Classification
4.2 Testing and Results
5 Conclusion
References
4 Enhancing the Management of Traffic Sequence Following Departure Trajectories
Abstract
1 Introduction
2 Problematic
3 State of Art
4 Modelization
5 Short Job First (SJF) Scheduling Concept
6 Computational Study and Data
7 Conclusion
References
5 A Multiagent and Machine Learning Based Denial of Service Intrusion Detection System for Drone Networks
Abstract
1 Introduction
2 Background of the Study
2.1 Drone Definition
2.2 Fleet of Drones
2.3 Geographic Data
2.4 Security Aspect of a Fleet of Drones
2.5 Computer Security
2.6 Network Intrusion Detection System—NIDS
3 State of the Art
3.1 Security Aspect in UAV Networks
3.2 DoS Attacks Targeting a Drone
3.3 Discussion
4 Proposed Approach
4.1 General View
4.2 Inspirational Model
4.3 Proposed Model
4.4 Components of the Proposed System
4.5 Principle of Operation of the Proposed NIDS Model
4.6 Intrusion Detection Mechanism (Micro Level of Detection)
5 Tests and Results
5.1 Training Dataset
5.2 Machine Learning Techniques
5.3 Recap of the Achieved Results
6 Conclusion and Perspectives
References
6 Toward a Deep Learning Approach for Automatic Semantic Segmentation of 3D Lidar Point Clouds in Urban Areas
Abstract
1 Introduction
2 Automatic Segmentation of 3D Point Clouds
2.1 Direct Approaches
2.2 Derived Product Based Approaches
2.3 Hybrid Approaches
2.4 Summary
3 Contribution of DL to Semantic Segmentation
4 Discussion
5 Our Approach
5.1 Methodology
5.2 Preliminary Segmentation
5.2.1 Data
5.2.2 Results
6 Summary
7 Conclusion
References
7 Artificial and Geospatial Intelligence Driven Digital Twins’ Architecture Development Against the Worldwide Twin Crisis Caused by COVID-19
Abstract
1 Introduction
2 The World Against a Triple Crisis
2.1 MENA Zone Countries on the Loop
2.2 Contextual Analysis of Pandemic Evolution: Focus on Spatial and Temporal Variations
2.3 Spatial and Temporal Analysis of Pandemic Evolution and Its Impacts Within Moroccan Territory
2.3.1 Spatiotemporal and Economic Analysis of the Pandemic
3 Global Efforts up to Date to Counter the Triple Crisis—Focus on Advanced Modelling, Dynamic Simulation and Artificial Intelligence AI
3.1 Beyond COVID-19 Towards a New Organizational Model for Value Chain Resiliency
4 Location Intelligence and Digital Twins State of the Art of Existing Solutions and Their Potential for Economic, Social and Industrial Resiliency
4.1 Digital Twins’ Concept, Applications and Paradigms for Implementation
4.2 Geospatial and Business Intelligence for Context Aware and Smart DT
5 DT-Geo-BI Platform for Smart and Resilient Supply Chains
6 Application Use Cases and Discussion
6.1 Use Case—Occupational Health and Safety System Resiliency Within Facial Protection Masks Value Chain
7 Conclusion and Future Research Axes
Acknowledgements
References
Remote Sensing and Artificiel Intelligence
8 Opportunities for Artificial Intelligence in Precision Agriculture Using Satellite Remote Sensing
Abstract
1 Introduction
2 Precision Agriculture
3 The Potential of Artificial Intelligence in Precision Agriculture
4 Artificial Intelligence Applied to Precision Agriculture
4.1 Machine Learning
4.2 Classification Algorithms
4.3 Regression Algorithms
4.4 Deep Learning
4.5 Genetic Algorithms
5 Overall Review of AI Application in Precision Agriculture
6 Opportunities, Challenges, and Future Trends
6.1 The Necessity of Expertise and Variability for Standardization of Treatments
6.2 The Complexity of Treatments and Costs
6.3 Data Availability
6.4 Algorithms
7 Conclusion
References
9 Monitoring Land Productivity Trends in Souss-Massa Region Using Landsat Time Series Data to Support SDG Target 15.3
Abstract
1 Introduction
2 Study Area
3 Data and Methods
3.1 Landsat Time Series and Composite Data
3.2 Calculate Sensor Calibration Coefficients
3.3 Correction of Errors Due to Clouds, Cloud Shadows, and Haze Present on the Images
3.4 Land Cover Data
3.5 Methodology
3.6 Calculating Productivity Metrics
3.6.1 Trend
3.6.2 State
3.6.3 Performance
3.7 Aggregation of the Productivity Sub-indicators
4 Results
4.1 Trend
4.2 State
4.3 Performance
4.4 Combination of Productivity Indicators
5 Discussion and Conclusions
References
10 Subimages-Based Approach for Landslide Susceptibility Mapping Using Convolutional Neural Network
1 Introduction
2 Materials and Methods
2.1 Data
2.2 Geospatial Modeling Tools
2.3 Conditioning Factors
2.4 Deep Learning
3 Results and Discussion
3.1 Model Parameters
3.2 Susceptibility Index
3.3 Discussion
4 Conclusion
References
11 Lithological Mapping for a Semi-arid Area Using GEOBIA and PBIA Machine Learning Approaches with Sentinel-2 Imagery: Case Study of Skhour Rehamna, Morocco
Abstract
1 Introduction
2 Location and Geological Settings of the Study Area
3 Materials and Methods
3.1 EO Datasets Properties and Pre-processing
3.2 Methodology
3.3 Results and Discussion
4 Conclusions
References
12 Optimization of Object-Based Image Analysis with Genetic Programming to Generate Explicit Knowledge from WorldView-2 Data for Urban Mapping
Abstract
1 Introduction
2 Genetic Programming
3 Methodology and Experiments
3.1 Study Area
3.2 Preprocessing of Input Data
3.3 Feature Extraction
3.4 Feature Selection
3.5 Generating Classification Rules
4 Results and Discussion
4.1 Validation Metrics
4.2 Statistical Metrics
5 Conclusion
References
13 Machine Learning and Remote Sensing in Mapping and Estimating Rosemary Cover Biomass
Abstract
1 Introduction
2 Materials and Methods
2.1 Sentinel 2 MSI Images
2.2 Spectral Indices
2.3 Random Forest Model
2.4 Biomass Calculation
2.4.1 Tuff Weight
2.4.2 Number of Tuffs Per Parcel
2.4.3 Wet and Dry Biomass Estimation
2.5 Results
2.5.1 Spectral Indices
2.5.2 Random Forest Classifier
2.5.3 Dry Biomass Estimation
2.6 Conclusion
Acknowledgments
References