Field Crop 2025, Vol.8, No.4, 204-212 http://cropscipublisher.com/index.php/fc 204 Research Insight Open Access Integrating UAV-Based Remote Sensing and Machine Learning to Monitor Rice Growth in Large-Scale Fields Deshan Huang, Yuandong Hong, Jianquan Li Hier Rice Research Center, Hainan Institute of Tropical Agricultural Resources, Sanya, 572025, Hainan, China Corresponding email: jianquan.li@hitar.org Field Crop, 2025, Vol.8, No.4 doi: 10.5376/fc.2025.08.0020 Received: 17 Jun., 2025 Accepted: 29 Jul., 2025 Published: 20 Aug., 2025 Copyright © 2025 Huang et al., This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Preferred citation for this article: Huang D.S., Hong Y.D., and Li J.Q., 2025, Integrating UAV-based remote sensing and machine learning to monitor rice growth in large-scale fields, Field Crop, 8(4): 204-212 (doi: 10.5376/fc.2025.08.0020) Abstract As one of the most important food crops in the world, the production level of rice is directly related to food security and sustainable agricultural development. The growth monitoring of rice in the field environment is confronted with challenges such as wide distribution of fields, significant environmental differences and complex growth processes. Unmanned aerial vehicle (UAV) remote sensing technology, with its high spatial resolution and flexibility, provides a new approach for obtaining growth parameters of rice populations. Meanwhile, machine learning methods have significant advantages in multi-source data processing, feature extraction and pattern recognition. This study explores the integrated framework of unmanned aerial vehicle (UAV) remote sensing and machine learning, summarizes the application characteristics of various sensors (RGB, multispectral, hyperspectral, and thermal infrared), assesses the applicability of vegetation index, canopy structure parameters, and physiological and ecological indicators in rice growth monitoring, and analyzes the performance of models such as random forest, support vector machine, XGBoost, and deep learning The application potential of this technology in the prediction of rice yield and disease monitoring in the Yangtze River Basin of China and Southeast Asia was demonstrated through case studies. The research results show that the combination of UAV and machine learning can effectively achieve precise monitoring of large-scale rice growth, which is of great significance to the development of precision agriculture and smart agriculture. This study aims to construct a monitoring framework integrating unmanned aerial vehicle (UAV) remote sensing and machine learning to achieve dynamic, precise and large-scale assessment of the growth status of rice in the field. Keywords Rice; Unmanned aerial vehicle remote sensing; Machine learning; Growth monitoring; Precision agriculture 1 Introduction Rice (Oryza sativa L.) is a very important crop in the world. It provides staple food for more than half of the global population and is also one of the most important grain crops in China. Maintaining high rice yields is crucial for addressing issues such as population growth, climate change, and resource shortages (Yuan et al., 2024; Chen et al., 2025). To achieve this, it is necessary to monitor the growth of rice on a large scale. Only in this way can it be managed better, yields be increased, and sustainable agricultural development be promoted at the same time. In the past, people mainly monitored the rice in the fields through manual investigation and ground observation. These methods are both slow and tiring, and it is difficult to achieve precision in terms of time and space, often failing to meet the needs of quick decision-making. Although satellite remote sensing can view a large area, its resolution is relatively low and it is easily affected by clouds and the atmosphere, thus failing to obtain details of the fields (Chen et al., 2024). In contrast, unmanned aerial vehicle (UAV) remote sensing has more advantages. It can quickly and flexibly obtain high-resolution data without damaging the plants. Coupled with machine learning algorithms, unmanned aerial vehicles can estimate biomass, nitrogen content, leaf area index (LAI), and yield relatively accurately even in complex field environments (Liu et al., 2023; Sarkar et al., 2023; Du et al., 2024; Ko et al., 2024). When dealing with images, methods such as random forests, support vector machines and deep neural networks usually perform better than traditional regression methods. They can extract more useful information from spectral and texture features (Zha et al., 2020; Bin et al., 2023; Wang et al., 2023).
RkJQdWJsaXNoZXIy MjQ4ODYzNA==