Early detection of crops with UAVs utilizing deep learning techniques: A Review

Abstract

Various biotic and abiotic stressors are applied to crops. The biotic factors in this situation are weeds, bacteria, fungi, and pests, while the abiotic stresses are soil pH, water salinity, temperature, humidity, and weather. Unmanned Aerial Vehicles (UAVs) provide a practical way to diagnose stress early and treat it quickly, increasing the yield of crops. Applications of deep learning and computer vision during plant surveillance can help achieve these goals more quickly and precisely by collecting data and comparing it to an existing dataset.Multispectral sensors employed on gimbles can be used in the future, which have been verified and trained with a range of datasets, and should be used for UAV-based crop detection. Because of their ability to measure soil fertility, identify crop diseases, increase productivity, and uncover a range of crop-affecting factors, drones are helpful in the management of natural resources and agricultural operations. UAVs can identify areas that need more attention. In the agricultural sector, unmanned aerial vehicles can be used to reduce the time and potentially hazardous effects of fertilizer and pesticide spraying. This study provides a brief overview of the usage of UAVs for agricultural surveillance.