Soil moisture or moisture content is a fundamental constituent of the hydrological system of the Earth and its ecological systems, playing a pivotal role in the productivity of agricultural produce, climate modeling, and water resource management. This review comprehensively examines conventional and advanced approaches for estimation or measuring of soil moisture, including in situ methods, remote sensing technologies, UAV-based monitoring, and machine learning-driven models. Emphasis is primarily on the evolution of soil moisture measurement from destructive gravimetric techniques to non-invasive, high-resolution sensing systems. The paper emphasizes how machine learning modules like Random Forest models, support vector machines, and AI-based neural networks are becoming more and more popular for modeling intricate soil moisture dynamics with data from several sources. A bibliometric analysis further underscores the research trends and identifies key contributors, regions, and technologies in this domain. The findings advocate for the integration of physics-based understanding, sensor technologies, and data-driven approaches to enhance prediction accuracy, spatiotemporal coverage, and decision-making capabilities.

Crop production heavily relies on precise management of crops and soil properties. Factors like temperature, structure and resistivity of crops and soil play crucial roles in determining agricultural output. Drone technology has emerged as the most effective solution, facilitating tasks such as chemical spraying, fertiliser application, irrigation management, weed detection, and soil analysis. This technology enables early detection of crop and soil conditions to implement timely remedies that enhance productivity. Drones reduce laborious tasks, lower input costs, and save operational time. Drones are categorised into some groups such as fixed wing, multi rotor, hybrid, remote piloted, autonomous, surveillance, spraying and payload specific drones. Equipped with sensors such as red‐blue‐green (RGB), multispectral, hyperspectral, thermal, spectrophotometer, radiometer and light detection and ranging (LiDAR), drones capture real‐time data of crops and farmland. These images undergo processing through advanced data analytics such as deep learning and machine learning. Vegetation indices such as normalised difference vegetation index, excess greenness index, normalised difference index and ratio of vegetation index etc. are employed to differentiate objects and distinguish crops from soil backgrounds. This review discusses various drone types, their components including sensors, methods for data processing, drone applications, their benefits, existing challenges, and considerations for future advancements.