<![CDATA[Artificial intelligence-based unmanned aerial vehicle UAV systems offer promising capabilities for oil palm plantation monitoring and quality assessment. However, most existing studies focus mainly on ripeness classification using RGB imagery, while integrated oil quality prediction, multimodal sensing, explainable artificial intelligence XAI, and real-time deployment remain limited. This study conducts a systematic literature review of AI-based UAV image processing technologies for oil palm ripeness detection and oil quality assessment using the PRISMA 2020 framework. We analyzed 45 peer-reviewed journal articles published between 2019 and 2025. The review indicates that convolutional neural network-based models are the most widely used for ripeness classification, while YOLO-based lightweight detectors are promising for real-time deployment. Multimodal sensing using RGB, multispectral, hyperspectral, thermal, and near-infrared data improves robustness for estimating oil quality-related variables such as oil extraction rate, moisture content, and free fatty acids. Key challenges include limited datasets, environmental variability, computational constraints, and the low adoption of XAI. Based on these findings, the study proposes an integrated AI-UAV smart plantation framework for sustainable precision agriculture in the oil palm industry.]]>
Copyrights © 2026
