IoT and AI in Agriculture: Self- sufficiency in Food Production to Achieve Society 5.0 and SDG's Globally

Foreword
Preface
Acknowledgments
Contents
Chapter 1: IoT x AI: Introducing Agricultural Innovation for Global Food Production
	1.1 Introduction
	1.2 Key Factors for Plant Growth and Agricultural Production
		1.2.1 Source of Light for Indoor Farming Systems
		1.2.2 IoT-Based Precision Irrigation Systems
		1.2.3 IoT-Based Water Purification
	1.3 Artificial Intelligence for Smart Agriculture
	1.4 Agricultural Machinery Automation
		1.4.1 Farm Automation Technology
		1.4.2 Agricultural Robot Navigation System
		1.4.3 Automation in Orchard Management
	1.5 Conclusion
	References
Chapter 2: Strategic Short Note: Transforming Controlled Environment Plant Production Toward Circular Bioeconomy Systems
	2.1 Introduction
	2.2 Circular CEPPS
	2.3 Closing Remarks
	References
Chapter 3: Artificial Lighting Systems for Plant Growth and Development in Indoor Farming
	3.1 Introduction
	3.2 Light for Plant Growth
	3.3 Light Quantity
		3.3.1 Plant Photosynthesis in Response to Light
	3.4 Light Quality
		3.4.1 Light Energy Use Efficiency of Lamps (LUEL) and Plant Community (LUEP)
	3.5 Light Duration or Photoperiod
		3.5.1 Daily Light Integral (DLI)
	3.6 Artificial Lights for Plants Growth
		3.6.1 Incandescent Lamps
		3.6.2 Fluorescent Lamps
		3.6.3 High Intensity Discharged (HID) Lamps
		3.6.4 Light Emitting Diodes (LED)
	3.7 Effect of Using Artificial Light on Plants Grown Indoor
	3.8 Conclusion
	References
Chapter 4: An IoT-Based Precision Irrigation System to Optimize Plant Water Requirements for Indoor and Outdoor Farming Systems
	4.1 Introduction
	4.2 Precision Irrigation Management
		4.2.1 IoT Technologies for Precision Irrigation Systems
			4.2.1.1 IoT Networking Backbone
			4.2.1.2 IoT: Irrigation Control System
				4.2.1.2.1 Fuzzy Logic-Based Control System
				4.2.1.2.2 Artificial Neural Network-Based Control System
				4.2.1.2.3 Hybrid Control System
		4.2.2 Indoor Precision Watering Management
			4.2.2.1 Soil Moisture-Based Scheduling
			4.2.2.2 Plant Water Status-Based Scheduling
		4.2.3 Outdoor Precision Irrigation Management
			4.2.3.1 IoT-Based Irrigation Scheduling
				4.2.3.1.1 Soil Moisture-Based Irrigation Scheduling
				4.2.3.1.2 Weather-Based Irrigation Scheduling
				4.2.3.1.3 Plant-Based Irrigation Scheduling
	4.3 Discussion
	4.4 Concluding Remarks
	References
Chapter 5: Strategic Short Note: Artificial Intelligence and Internet of Things: Application in Urban Water Management
	5.1 Introduction
	5.2 Methods
	5.3 Results
	5.4 Conclusion
	References
Chapter 6: Purification of Agricultural Polluted Water Using Solar Distillation and Hot Water Producing with Continuous Monito...
	6.1 Introduction
	6.2 The Architecture of the Proposed IoT-Based Solar Water Distillation and Hot Water System
	6.3 Basic Architecture of an IoT-Based Water Purification System
	6.4 Water Purification Methods and the Possibility of Using IoT
	6.5 Solar Water Distillation and Potential for Improvement with the Latest Innovations
		6.5.1 Active and Passive Solar Distillation
		6.5.2 Possible Innovations to Improve Vapor Generation in Solar Stills
	6.6 Solar Water Heating Systems and Data Monitoring
	6.7 IoT-Based Solar Water Distillation and Hot Water System
		6.7.1 Study Conducted to Test the Performance of Solar Stills Under Different Improvement Strategies
		6.7.2 Performance Evaluation of the Solar Still
		6.7.3 Water Quality Results
	6.8 Conclusions
	References
Chapter 7: Long Range Wide Area Network (LoRaWAN) for Oil Palm Soil Monitoring
	7.1 Introduction
	7.2 Internet of Things (IoT) in Agriculture
	7.3 Wireless Sensor Network in Agriculture
	7.4 Soil Electrical Conductivity (EC) and pH in Oil Palm
	7.5 LoRaWAN System Design for Soil EC and pH Monitoring
	7.6 Signal Propagation Tests
		7.6.1 Signal Propagation Test in a Young Oil Palm Plantation
		7.6.2 Signal Propagation Test in an Oil Palm Nursery
		7.6.3 Signal Propagation Test in an Urban Area
	7.7 Calibration of EC and pH Sensors
	7.8 Soil EC and pH Measurement Test
	7.9 Conclusion
	7.10 Recommendation
	References
Chapter 8: Strategic Short Note: Application of Smart Machine Vision in Agriculture, Forestry, Fishery, and Animal Husbandry
	8.1 Introduction
	8.2 Tasks of Smart Machine Vision
	8.3 The Components of Smart Machine Vision
	8.4 Examples of Smart Machine Vision in Agriculture, Forestry, Fishery, and Animal Husbandry
	8.5 Conclusion
	References
Chapter 9: Artificial Intelligence in Agriculture: Commitment to Establish Society 5.0: An Analytical Concepts Mapping for Dee...
	9.1 Introduction
	9.2 AI Mapping Concept
	9.3 Deep Learning (DL) and Neural Networks (NNs)
		9.3.1 Principles of the ANN Learning Process
	9.4 DL at the Edge with CNNs
	9.5 Deep Learning Algorithms for Object Detection
	9.6 Conclusions: DL Serving a New Agriculture Revolution
	References
Chapter 10: Potentials of Deep Learning Frameworks for Tree Trunk Detection in Orchard to Enable Autonomous Navigation System
	10.1 Introduction
	10.2 Materials and Methods
		10.2.1 Field Data Collection
		10.2.2 Data Preparation
			10.2.2.1 Image Frames from Videos
			10.2.2.2 Labeling
			10.2.2.3 Data Augmentation
			10.2.2.4 Data Splitting
		10.2.3 Training Model Structure
			10.2.3.1 Faster R-CNN (Faster Region Based Convolutional Neural Networks)
				10.2.3.1.1 Convolutional Layers
				10.2.3.1.2 RPN
				10.2.3.1.3 ROI Pooling
				10.2.3.1.4 Classification
			10.2.3.2 YOLO (You Only Look Once)
			10.2.3.3 CenterNet
		10.2.4 Training Platform and Validation
		10.2.5 Model Testing
	10.3 Results
		10.3.1 Faster R-CNN Testing
		10.3.2 YOLO Testing
		10.3.3 CenterNet Testing
	10.4 Discussion
	10.5 Conclusion
	References
Chapter 11: Real-Time Pear Fruit Detection and Counting Using YOLOv4 Models and Deep SORT
	11.1 Introduction
	11.2 Related Works
		11.2.1 You Only Look Once (YOLO)
		11.2.2 Simple Online and Real Time Tracking with Convolutional Neural Networks (CNNs)
		11.2.3 Fruit Detection Using YOLO
		11.2.4 Real-Time Fruit Counting Using YOLO and an Object Tracking Algorithm
	11.3 Materials and Methods
		11.3.1 Field Data Collection
		11.3.2 Data Preparation
			11.3.2.1 Videos Were Converted into Image Frames
			11.3.2.2 Labelling
			11.3.2.3 Data Augmentation
		11.3.3 Data Splitting
		11.3.4 Setting the Target Metric
		11.3.5 Evaluation Metrics for the Detection
		11.3.6 Components of the YOLOv4 Models
			11.3.6.1 Cross-Stage Partial (CSP) Connection
			11.3.6.2 CSPDarknet53: YOLOv4 and YOLOv4-CSP´s Backbone
			11.3.6.3 YOLOv4-tiny´s Backbone: CSPOSANet
			11.3.6.4 Why Were Leaky Rectified Linear Unit and Mish Used as the Activation Functions for the YOLOv4 Models?
			11.3.6.5 YOLOv4´s Neck: Path Aggregation Network (PANet)
			11.3.6.6 YOLOv4´s Plug-In Module: Spatial Pyramid Pooling (SPP)
		11.3.7 Training, Validation, and Optimization
			11.3.7.1 Stage-1 Training
			11.3.7.2 Hyperparameters
			11.3.7.3 Stage-2 Training
			11.3.7.4 Error Analysis
		11.3.8 Model Comparison
		11.3.9 Pear Counting Using the Selected YOLOv4 Model and Deep SORT
		11.3.10 Evaluation Metrics for the Pear Counting
	11.4 Results and Discussion
		11.4.1 Training Details
		11.4.2 Model Performance Comparison
		11.4.3 Speed-Accuracy Tradeoff in the YOLOv4 Models
		11.4.4 Average Precision at Different Thresholds
		11.4.5 FLOPS Analysis
		11.4.6 YOLOv4 Models on Illumination and Occlusion Challenges
		11.4.7 Comparison of the Pear Counting Methods
		11.4.8 Breakdown of the False Negative Counts in the ROI Line-Based Counting
	11.5 Conclusions
	Appendix
	References
Chapter 12: Pear Recognition System in an Orchard from 3D Stereo Camera Datasets Using Deep Learning Algorithms
	12.1 Introduction
	12.2 Materials and Methods
		12.2.1 Field Data Collection
		12.2.2 Instance Segmentation
		12.2.3 Mask R-CNN
		12.2.4 ZED AI Stereo Camera
		12.2.5 Data Preparation
			12.2.5.1 Deep Learning Environment
			12.2.5.2 Video to Image Conversion
			12.2.5.3 Image Annotation
		12.2.6 Data Splitting
		12.2.7 Training Process of Mask R-CNN
			12.2.7.1 Feature Extraction (Backbone: ResNet101 + FPN)
			12.2.7.2 Region Proposal Network (RPN)
			12.2.7.3 ROIs and ROI-Align
			12.2.7.4 Mask RCNN for Classification and Regression
			12.2.7.5 Loss Function
			12.2.7.6 Model Metrics Function
	12.3 Results
		12.3.1 Training Details
		12.3.2 Evaluation of Model Metrics
		12.3.3 Evaluation of Model Effectiveness
	12.4 Discussion
	12.5 Conclusion
	References
Chapter 13: Thermal Imaging and Deep Learning Object Detection Algorithms for Early Embryo Detection: A Methodology Developmen...
	13.1 Introduction
	13.2 Materials and Methods
		13.2.1 Thermal Imaging
			13.2.1.1 Transmittance (τ)
			13.2.1.2 Emissivity (ε)
			13.2.1.3 Reflectance (ρ)
		13.2.2 Experimental Environment
		13.2.3 Thermal Image Acquisition and Radiometric Corrections
		13.2.4 Deep Learning Algorithms and Analysis Environment
			13.2.4.1 Models Training
			13.2.4.2 Data Labeling
			13.2.4.3 Data Augmentation
			13.2.4.4 Model Evaluation
	13.3 Results
		13.3.1 Thermal Features of Incubating Eggs
		13.3.2 Training Results
	13.4 Discussion
	13.5 Conclusions
	References
Chapter 14: Strategic Short Note: Intelligent Sensing and Robotic Picking of Kiwifruit in Orchard
	14.1 Introduction
	14.2 Intelligent Sensing of Kiwifruit
	14.3 Nondestructive Picking of Fruit
	14.4 Kiwifruit Picking Robot
	14.5 Conclusions
	References
Chapter 15: Low-Cost Automatic Machinery Development to Increase Timeliness and Efficiency of Operation for Small-Scale Farmer...
	15.1 Introduction
	15.2 Current Agricultural Trends
		15.2.1 Control and Navigation System
		15.2.2 Vehicle Motion Models
		15.2.3 Navigation Planner
		15.2.4 Steering Controllers
		15.2.5 Field Sensing, Recognition, and Sensor Data Fusion
		15.2.6 Variable-Rate Technologies
		15.2.7 Communication Protocols
	15.3 Levels of Automation in Farm Machinery
		15.3.1 Level 0: No Automation
			15.3.1.1 Transformation of Automation on a Power Tiller
		15.3.2 Level 1: Assisted Automation
			15.3.2.1 Transformation of Automation for Seed/Fertilizer Broadcasting Device
		15.3.3 Level-2: Partial Automation
			15.3.3.1 Transformation of Partial Automation
		15.3.4 Level 3: Conditional Automation
			15.3.4.1 Transformation of Conditional Automation
		15.3.5 Level 4: High Automation
			15.3.5.1 Transformation of Automation for Machinery
		15.3.6 Level 5: Full Automation
			15.3.6.1 Transformation of Automation Systems
	15.4 Discussion
	15.5 Conclusion
	References
Chapter 16: Vision-Based Leader Vehicle Trajectory Tracking for Multiple Agricultural Vehicles
	16.1 Introduction
	16.2 Materials and Methods
		16.2.1 Leader-Follower Relative Position and Camera-Marker Sensing System
			16.2.1.1 Camera Servo System
			16.2.1.2 Marker Detection
			16.2.1.3 Marker Positioning
			16.2.1.4 Offset of Roll Angle between Camera and Marker
			16.2.1.5 Transformation of Coordinates and Relative Positioning of the Marker
		16.2.2 Camera Vision Data Estimation and Smoothing
		16.2.3 Design of Control Law for the Leader Trajectory Tracking of Follower Vehicle
	16.3 Field Experiments
	16.4 Results and Discussion
		16.4.1 Evaluation of Camera-Marker Observation System
		16.4.2 Tracking Performance
	16.5 Discussion
	16.6 Conclusions
	References
Chapter 17: Autonomous Robots in Orchard Management: Present Status and Future Trends
	17.1 Introduction
	17.2 Vision Systems for Autonomous Orchard Robots
	17.3 Autonomous Robotic Pruning in the Orchards
	17.4 Pollination of Orchards Crops Using Autonomous Robots
	17.5 Use of Fertilizer and Liquid Chemical Application Autonomous Robots in Orchards
	17.6 Autonomous Robots for Harvesting Fruits in Orchards
	17.7 Discussion
	17.8 Conclusions
	References
Chapter 18: Strategic Short Note: Comparing Soil Moisture Retrieval from Water Cloud Model and Neural Network Using PALSAR-2 f...
	18.1 Introduction
	18.2 Retrieval of Soil Moisture Content in Oil Palm Fields
	18.3 Conclusion
	References
Chapter 19: Development of a Recognition System for Spraying Areas from Unmanned Aerial Vehicles Using a Machine Learning Appr...
	19.1 Introduction
	19.2 Materials and Methods
		19.2.1 Mutual Subspace Method (MSM)
		19.2.2 Research Design for Classifiers and MSM
		19.2.3 Field Experiment for Training and Testing with Datasets
		19.2.4 Offline Recognition System
		19.2.5 Online Recognition System
	19.3 Results
		19.3.1 Offline Recognition Performance
		19.3.2 Online Recognition Performance
	19.4 Discussion
	19.5 Conclusion
	References
Chapter 20: Basal Stem Rot Disease Classification by Machine Learning Using Thermal Images and an Imbalanced Data Approach
	20.1 Introduction
	20.2 Basal Stem Rot (BSR)
	20.3 Detection of BSR Disease
		20.3.1 Remote Sensing Techniques for G. boninense Disease Detection
		20.3.2 Detection of G. boninense Using Thermal Imaging
	20.4 Machine Learning in Crop Disease
		20.4.1 Machine Learning in BSR Disease Detection
	20.5 Imbalanced Data Approach
		20.5.1 Data-Level Approaches
			20.5.1.1 Under-Sampling
			20.5.1.2 Oversampling
			20.5.1.3 Synthetic Minority Oversampling Technique (SMOTE)
	20.6 Experimental Methodology
		20.6.1 Thermal Data Acquisition
			20.6.1.1 Emissivity Measurement
			20.6.1.2 Reflected Apparent Temperature (RAT)
			20.6.1.3 Atmospheric Temperature and Humidity
			20.6.1.4 Object-to-Camera Distance
		20.6.2 Pre-processing of Thermal Images
		20.6.3 Thermal Image Feature Extraction
		20.6.4 Statistical Analysis
		20.6.5 Machine Learning Approach
		20.6.6 Imbalance Data Approach
	20.7 Experimental Evaluation
		20.7.1 Time Session Selection
		20.7.2 Selection of Feature Temperature
		20.7.3 Classification Analysis of Feature Temperature
		20.7.4 The Effect of Classifiers on Model Performance
		20.7.5 The Effect of Data Imbalance on Classification
	20.8 Conclusions
	References
Chapter 21: Early Detection of Plant Disease Infection Using Hyperspectral Data and Machine Learning
	21.1 Basal Stem Rot (BSR) Disease due to G. boninense Infection
	21.2 Hyperspectral Imaging
	21.3 Machine Learning
	21.4 Research Design
		21.4.1 Research Area
		21.4.2 Preparation of Samples
			21.4.2.1 Artificial Inoculation
			21.4.2.2 Polymerase Chain Reaction (PCR)
		21.4.3 Hyperspectral Imaging
			21.4.3.1 Image Acquisition
			21.4.3.2 Spectral Extraction
			21.4.3.3 Significant Bands for BSR Detection
	21.5 BSR Detection
		21.5.1 BSR Detection Using SVM
		21.5.2 BSR Detection Using Various Types of ML
		21.5.3 BSR Detection Using SVM and a Small Number of Bands
	21.6 Conclusion
	References
Chapter 22: Strategic Short Note: Development of an Automated Speed Sprayer for Apple Orchards in Japan
	22.1 Introduction
		22.1.1 Apple Production and Pesticide Spraying
	22.2 Past Efforts to Automate SS and Development Goals
	22.3 GNSS Application
	22.4 ArUco Markers and GNSS Application
	22.5 LiDAR Application
	22.6 Future Tasks
	Reference
Chapter 23: The Spectrum of Autonomous Machinery Development to Increase Agricultural Productivity for Achieving Society 5.0 i...
	23.1 Introduction
	23.2 Autonomous Machinery: Japanese Spectrum and Current State of Development
	23.3 Solution Developments
	23.4 Industrial Anticipation and Outlook
	23.5 Market Commercialization with Geographic Application Targets
	23.6 Conclusions
	References