Project Number: 2090-21000-037-002-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: May 15, 2023
End Date: May 14, 2024
Field evaluation of 734 pea (Pisum sativum L.) accessions focused on yield component traits using phenomic tools.
Field Experiment: A field study of 734 pea accessions planted using an augmented design with repeated 3 check cultivars for a total of 840 plots. Yield component trait data will be collected using UAV and ground robot mounted RGB and multispectral cameras. Traits to be collected include emergence, early vigor, stand count, days to row closure, days to 10% flower, plant health, pod density, plant height, biomass and maturity. The RGB and multispectral images from UAS will be preprocessed using the Pix4Dmapper photogrammetry software (Pix4D S.A., Lausanne, Switzerland) to derive a point cloud data and an orthomosaic image, respectively. The software automatically utilized the scene illumination, reference panel, and sensor specifications to improve the orthomosaic images’ radiometric quality. These two raw data types will be processed to extract the digital traits, such as vegetation indices commonly used in agriculture (https://www.l3harrisgeospatial.com/docs/vegetationindices.html) using orthomosaic imagery and traits such as plant height and canopy volume using point cloud. These processes will be similar to those described in Valencia-Ortiz et al. (2021) and Sangjan et al.(2022b), respectively. Some of the vegetation indices can include normalized difference vegetation index, green normalized index vegetation index, normalized difference red edge index, simple ratio index, etc. Other approaches available for analyzing legume data from literature will also be adapted as appropriate (Zhang et al., 2021b, Bari et al., 2022). Ground truth data to be collected includes emergence, early vigor, stand count, days to row closure, days to 10% flower, plant health, pod density, plant height, biomass and maturity.