Using Uav Imagery for Crop Analytics.

UAV imagery offers an efficient and effective way to survey small to large scale farming operations to assess overall crop health. This paper will address how to apply several remote sensing techniques to extract actionable information for crop management from UAV imagery. UAV imagery offers higher spatial, spectral and temporal resolution at a lower cost when compared to imagery collected from traditional manned aircraft or satellite sensors. First, we will introduce a novel approach for automatically counting crops from UAV imagery. The crop counting algorithm outputs the number, position and size of individual plants in a field, which can lead to increased accuracy of yield predictions. Next, we will look at applying geospatial statistical analytics for monitoring vegetation health as well as hot spot analysis to determine which areas of a field require attention. The specific analytics used will be dependent upon the spectral resolution of the sensor mounted on the UAV. Finally, we will address monitoring change in crop health throughout the growing season by applying change detection techniques to multispectral, multi-temporal imagery collected from a UAV.