Optical and Thermal IR Sensing of Crop Condition Using Multiple View Angles.

Optimal management of nitrogen and water conserves resources without risking food shortages, but optimal management is not easy. Collecting and analyzing soil or plant samples during the growing season may be too slow or too expensive to guide management decisions. Optical and thermal-IR sensors, which provide relative estimates of plant nitrogen or water status, are an increasingly affordable alternative. Before canopy closure, however, downward-pointing sensors (e.g., from a drone) view an unknown mix of soil and crop. Sensor data from multiple view angles might, in theory, distinguish % cover from leaf greenness and leaf temperature from soil temperature. A geometric computer model, inspired by Warren Wilson's 1963 analyses of point-quadrat data, was used to develop algorithms for analyzing multiple-sensor data. This approach was tested using field plots in Minnesota's Long-Term Agricultural Research Network.