A review of unmanned aerial vehicle based remote sensing and machine learning for cotton crop growth monitoring

• This work provides a novel review on machine learning for cotton crop monitoring. • We identified an urgent need for better data acquisition and management strategies. • Future work needs cost-effective systems for real-time data preprocessing. • Enhancing model scalability and precision in varied farms is critical. Cotton is one of the world’s most economically significant crops. Evaluating and monitoring cotton crop growth play vital roles in precision agriculture. Unmanned aerial vehicle (UAV) based remote sensing, when integrated with machine learning technologies, exhibits considerable promise for crop growth management. Despite these technologies’ substantial impact on cotton production, there exists a scarcity of consolidated information regarding various methods used. This paper offers a comprehensive review and analysis focused on methods for monitoring and evaluating cotton growth using UAV-based imagery combined with machine learning techniques. We synthesize the existing research from the past decade within this context, particularly discussing data acquisition strategies, preprocessing methods necessary for handling UAV-acquired images effectively, and a range of machine learning models applied. This investigation offers a comprehensive outlook that could guide future research efforts towards more efficient and sustainable agricultural practices in cotton production, leveraging state-of-the-art technology.