Sensing Technologies for Field and In-House Crop Production: Technology Review and Case Studies (Smart Agriculture, 7)

This book focuses on state-of-the-art sensing and automation technologies for field crops and in-house product production and provides a lot of innovative knowledge on image processing, AI algorithms and applications in agriculture, and robotics. This book provides undergraduate or graduate students with take-away knowledge for unmanned agricultural production, including but not limited to corn disease detection, wheat head detection and counting, and soil nutrient condition monitoring. The first three chapters focus on reviewing plant phenotyping sensing technology and robotics and soil nutrient monitoring, followed by in-house crop sensing robotics. Then two case studies on corn and the other two case studies on wheat are presented. Preface 6 Contents 8 1 A Review of Three-Dimensional Multispectral Imaging in Plant Phenotyping 9 1.1 Introduction 10 1.2 Three-Dimensional and Multispectral Imaging 12 1.2.1 Three-Dimensional Imaging 12 1.2.2 Multispectral Imaging 16 1.3 Three-Dimensional Multispectral Imaging Systems in Plant Phenotyping 17 1.3.1 System Structure 17 1.3.2 Factors Affecting Imaging 21 1.4 Commercial Three-Dimensional Multispectral Imaging Systems 22 1.5 Conclusion and Outlook 24 References 24 2 Recent Advances in Soil Nutrient Monitoring: A Review 27 2.1 Introduction 27 2.2 Basic Concepts of Soil Nutrients 29 2.3 Soil Nutrient Measurement Methods 30 2.3.1 Laboratory Analysis Methods 30 2.3.2 Spectroscopy Methods 32 2.3.3 Electrochemical Methods 37 2.4 Summary and Outlook 43 References 44 3 Plant Phenotyping Robot Platform 47 3.1 Overview of Crop Phenotyping Robots 47 3.2 Design Requirements of Crop Phenotyping Robot Platform 49 3.3 Research Status of Crop Phenotyping Robots 49 3.3.1 Robot Platform for Indoor Crop Phenotype 50 3.3.2 Field Crop Phenotyping Robot Platform 51 3.4 A Case Study of Tomato Plant Phenotype Inspection Robot 53 3.4.1 Hardware Components 53 3.4.2 Software Design 54 3.4.3 Extraction of Phenotype Information 57 References 59 4 Autonomous Crop Image Acquisition System Based on ROS System 61 4.1 Overview of the Indoor Crop Inspection System 62 4.1.1 Research Status of Indoor Crop Inspection Systems 62 4.1.2 Indoor Crop Inspection System Overall Design 63 4.2 System Hardware Design 65 4.2.1 Main Control Module 66 4.2.2 Motion Control Module 67 4.2.3 Image Acquisition Module 68 4.2.4 Real-Time Monitoring Module 69 4.3 Indoor Crop Inspection System Testing 73 4.3.1 Feasibility Testing of the System 73 4.3.2 System Stability Testing and Application 73 4.4 Extraction of Three-Dimensional Distribution of Potato Plant CWSI 74 4.4.1 Image Data Processing 74 4.4.2 Extraction of Crop Canopy CWSI 80 4.5 Conclusion 82 References 83 5 SeedingsNet: Field Wheat Seedling Density Detection Based on Deep Learning 85 5.1 Introduction 86 5.2 Materials and Methods 88 5.2.1 Experimental Field and Data Collection 88 5.2.2 Data Preprocessing 89 5.2.3 Detection Model 91 5.2.4 Performance Evaluation 93 5.3 Results 94 5.3.1 Results of Training Three Models 94 5.3.2 Comparison Results of Input Images with Different Processing Methods 94 5.4 Conclusion 95 References 96 6 Wheat Lodging Detection Using Smart Vision-Based Method 97 6.1 Introduction 97 6.1.1 Remote Sensing Role in Crop Lodging Detection 98 6.1.2 Lodging Severity 99 6.1.3 Changes in Wheat Lodging Over Time 99 6.1.4 Imbalance Challenge 100 6.2 Material and Methods 100 6.2.1 Image Data Gathering 100 6.2.2 Generating Auto Dataset 101 6.2.3 Feature Extraction 103 6.2.4 Changing Loss Function 103 6.2.5 Time-Dependent Modeling of Wheat Lodging 104 6.3 Results and Discussion 105 6.3.1 Classification Results of CNNs and Classic Classifiers 105 6.3.2 LSTM Based Model Classification Results 106 6.3.3 Changing the Loss Function in Wheat Lodging Classification 107 6.4 Conclusion 108 References 108 7 Design, Construction, and Experiment-Based Key Parameter Determination of Auto Maize Seed Placement System 111 7.1 Introduction 112 7.2 Materials and Methods 114 7.2.1 Introduction to the Auto Seeding Machine and Seed Placement System 114 7.2.2 The Overall Procedure of Key Parameter Determination for Seed Placement System 117 7.2.3 Bottom Tilt Angle of Seed Container 117 7.2.4 Brush Spinning and Linear Speed Configuration 120 7.2.5 Seed Dropping Height Study 121 7.3 Results and Discussion 124 7.3.1 Bottom Plate Angle Determination of Seed Container 124 7.3.2 Brush Spinning and Horizontal Speed Configuration 126 7.3.3 Seed Dropping Height 126 7.4 Conclusions 127 References 128 8 Development and Test of an Auto Seedling Detection System 129 8.1 Introduction 129 8.2 System Design 131 8.2.1 Auto Seedling Counting System Design 133 8.3 Materials and Methods 135 8.3.1 Image Acquisition 135 8.3.2 Color Images Correction 136 8.3.3 Visual Seeding Counting 137 8.3.4 Images Color Segmentation 137 8.3.5 Image-Block Seedlings Detection (IBSD) Algorithm 139 8.3.6 Evaluation 140 8.4 Results and Discussion 140 8.4.1 Effect of Seedling Growth days 140 8.4.2 Effect of Illumination Condition 141 8.5 Conclusion 142 References 143