Estimating Wheat Agronomic and Physiological Characteristics from Ground and Aerial RGB Imaging.

Although successful in developing high-yielding cultivars, current breeding programs use methods of selection that are slow, uneconomical, and inefficient. Breeding for targeted physiological traits was demonstrated superior to breeding for yield alone, but traditional selection methods were laborious and time consuming. Modern field evaluations can benefit from low-level aerial imaging platforms that are increasingly more available and affordable. Our objectives were to evaluate the efficacy of RGB-derived vegetation indices taken with an unmanned aerial vehicle (UAV) for estimation of yield, harvest index, and nitrogen use efficiency parameters; and to compare aerial images with traditional methods for estimating physiological and agronomic characteristics of wheat. An octocopter synchronized with a Sony Alpha 6000 digital camera were used to take images at an altitude of 10 to 30 m above the experimental plots in a waypoint navigation mode at GS 30, 45, 65, 75, and 85 at two locations in Virginia. Vegetation indices green area (GA, pixels with Hue from 60 to 120°), and greener area (GGA, pixels with Hue from 80 to 120), developed by Casadesus and Villegas (2012), were computed from .jpeg files with Image J software.  RGB images were also taken from the ground with a Samsung NX 300 camera. Normalized difference vegetation index (NDVI) readings were taken with a Trimble GreenSeeker handheld crop reflectance sensor; and the difference between canopy and air temperature was monitored with a handheld AGRI-THERM III infer-red thermometer. The correlations between aerial and ground indices and their potential use to estimate yield, harvest index, and nitrogen use efficiency will be discussed.