Field Crop 2025, Vol.8, No.4, 204-212 http://cropscipublisher.com/index.php/fc 205 This study will review the significance of rice in global food security and analyze the difficulties and the latest progress in large-scale monitoring. This paper mainly introduces the combination of unmanned aerial vehicle (UAV) remote sensing and machine learning, and proposes a research framework and objective for precise rice management. When these technologies are combined, they are expected to significantly enhance the accuracy and efficiency of monitoring and also expand the scope of application. Ultimately, these methods can promote sustainable food production and enhance the risk-resistance capacity of agriculture. This study aims to provide theoretical and practical references for establishing a precise and efficient rice monitoring system, and to promote the development of smart agriculture and digital villages. 2 UAV Remote Sensing Platforms and Data Acquisition Techniques 2.1 Advantages and limitations of UAVs in agricultural monitoring Drones have many advantages in agricultural monitoring. They can offer very high spatial and temporal resolution, fly flexibly, and collect detailed information at different growth stages of crops. Unmanned aerial vehicles (UAVs) collect data quickly, do not damage plants, and have relatively low costs. This is particularly valuable for large-scale rice fields and precise management (Figure 1) (Cen et al., 2019). However, drones also have some shortcomings. For instance, the battery life is limited, making it difficult to fly in windy and rainy weather, and it is also subject to regulatory control. To ensure data quality and coverage, meticulous flight planning and calibration are required (Zha et al., 2020). Meanwhile, unmanned aircraft will generate a large amount of image data, and processing and analyzing it requires strong computing resources and professional skills. Figure 1 Illustration of the UAV system and radiometric calibration targets (Adopted from Cen et al., 2019) 2.2 Applications of different sensor types (RGB, multispectral, hyperspectral, thermal infrared) Different sensors can all be installed on drones, each with its own advantages: RGB camera: Low price, simple operation, capable of taking high-resolution color images for evaluating canopy structure, color and height, but with limited spectral information (Yang et al., 2021). Multispectral camera: It can obtain data in multiple bands
RkJQdWJsaXNoZXIy MjQ4ODYzNA==