Sensors in Agriculture: Systems and Methods for Two Sensor Systems for Plant Phenotype Detection

Abstract

Sensors are increasingly being used in agricultural settings to provide data on the physical characteristics of plants under field conditions. Accurate data provides researchers and producers with the ability to make decisions with a high level of confidence. This work addresses two sensing systems for measuring important plant characteristics. The first system investigates accuracy differences between two unmanned aerial vehicle (UAV) camera calibration methods. The second system explores the development and testing of a novel in situ root imaging rhizotron. The UAV study compared autoexposure and fixed exposure radiometric calibration methods to a single calibrated manned aircraft image and to a ground target measured with a spectroradiometer. In a band by band comparison, the autoexposure method, which uses a pre-flight image of a single panel for calibration, produced almost twice as much radiometric error on average compared with fixed exposure using in-field targets for image calibration. When comparing the exposure methods using the Visible Atmospherically Resistant Index (VARI), the autoexposure method produced twice as much RMSE compared to the fixed exposure method. The study on the novel in situ root sensor developed a low field magnetic resonance imaging (LF-MRI) rhizotron. A scaled 8 cm bore model was designed, built and test across three types of soil, Weswood silt loam, Belk clay, and Houston black clay. The results demonstrated the viability of this technology to produce root information in clay soils. A 28 cm bore unit was designed, built and tested under field conditions. The resulting system provided root information and visualization of roots with 2-D projection images in a Weswood silt loam, and Belk clay both in situ and ex situ. In summary, (1) using a fixed exposure calibration method for UAV remote sensing improved accuracy in reflectance data, providing a better understanding of in-field plant conditions and better decision-making capability; and (2) the LF-MRI Rhizotron allowed visualization of plant roots in agricultural soils under field conditions. Both sensing systems and methods have the potential to be used as tools for improving crop production for researchers or growers.