Cloud Effects Correction Of Drone-Collected Crop Imagery For Precision Agriculture

The effect of cloud shadows on aerial imagery is an unsolved problem that has been identified by many operators of unmanned aerial vehicle (drone)-based remote sensing systems for agriculture. Shadows affect the ability to compare imagery, including multispectral and hyperspectral imagery (MSI and HSI), taken over periods of time, and the ability to quantify image data analytics, including NDVI and more sophisticated metrics for crop health, water stress, pigments, diseases, and soil nutrients. The overall objective of Phase I of this effort is to demonstrate component technologies needed to provide cloud effects correction in drone imagery, including development of a preliminary design for hardware and data post-processing software. Critical components include the collection and storage of skydata, algorithms to transform the data into quantifiable shadow and other cloud effects, and finally application of irradiance to cloud shadow correction in the down-looking imagery.