ABSTRACT
Unmanned aerial vehicles (UAVs) or flying drones and unmanned ground vehicles can capture a huge amount of spatial data from an agricultural field. However, efficiently processing such data is challenging. Computer vision algorithms such as deep learning-based methods could provide an efficient, cost-effective, flexible, and scalable remote-sensing solution for smart agriculture. These techniques could be used to cope with different agricultural challenges in many different areas, such as plant detection and classification, plant health assessment, smart pest and herb control, and field analysis and yield estimation. This chapter systematically reviewed different deep learning-based approaches, where UAV captured spatial data to solve these challenges. Here we also discuss technical backgrounds and future scopes.
