Transforming disease management through the use of unmanned aerial systems
Jan van Aardt: Rochester Institute of Technology, Chester F. Carlson Center for Imaging Science
<div>Relatively recent advances in unmanned aerial systems (UAS) technology, as well as miniaturization of complex remote sensing systems, have enabled new approaches to precision agriculture. For example, imaging spectroscopy (hyperspectral) and light detection and ranging (lidar) can be used for moisture stress assessment, nutrient mapping, and disease detection. We specifically are developing risk models for proactive management of white mold (<em>Sclerotinia sclerotiorum</em>) in snap beans. Prophylactic fungicide applications, applied when 10% of plants have at least one bloom, typically are used to prevent the mold. The goals are to identify spectral signatures of blooming onset, investigate spectral characteristics of white mold onset in the snap bean crop, and assess 3D lidar point clouds for structural inputs to risk models. The study area is located at the New York State Agricultural Experiment Station, Geneva, NY, USA, operated by Cornell University. A DJI Matrice-600 UAS, boasting a high spatial resolution color camera, a Headwall Photonics imaging spectrometer (272 bands; 400-1000 nm), and a Velodyne VLP-16 lidar system, is being used for this research. Our analysis approach involves a series of high frequency flights – centered around blooming and white mold onset - throughout the growing season. We will present the initial findings of this work, focusing on the need for proper calibration-to-reflectance of the imaging spectroscopy data, identifying an operational set of wavelengths from this spectrally oversampled imagery, and the benefit of fusing 3D lidar data with high fidelity spectral data for inclusion in the risk modeling effort. This UAS sensing approach hopefully one day can become standard practice.</div>
View Presentation |
