Application of UAV Remote Sensing in Crop Detection

Abstract

Unmanned Aerial Vehicle (UAV) remote sensing has become an important means of accurate detection of crops because of its high spatial-temporal resolution, flexibility, and cost control. This paper systematically summarizes the research progress of UAV Remote Sensing Application in crop detection. Firstly, the principles of mainstream sensors such as multispectral, hyperspectral, thermal infrared, visible light, and lidar and their application characteristics in agricultural monitoring are summarized. Secondly, focusing on the main tasks of pest identification, lodging monitoring, weed detection, and soil moisture monitoring, the retrieval methods based on different sensors are sorted out. On this basis, the advantages, disadvantages, and application scope of physical model and machine learning inversion models are compared. Research shows that the combination of multi-source sensor fusion and machine learning models significantly improves detection accuracy, but it still faces challenges such as poor model generalization ability, high sensor cost, and complex data processing. In the future, it should promote the coordinated development of UAV remote sensing, satellite and aerial remote sensing, and build a multi-scale, all-weather agricultural monitoring system.