Author
Jansky, Shelley | |
ROUSE, DOUG - University Of Wisconsin | |
GEVENS, AMANDA - University Of Wisconsin |
Submitted to: American Journal of Potato Research
Publication Type: Proceedings Publication Acceptance Date: 3/27/2014 Publication Date: 7/27/2014 Citation: Jansky, S.H., Rouse, D., Gevens, A. 2014. High throughput phenotyping using an unmanned aerial vehicle. American Journal of Potato Research. Paper No. G77. Interpretive Summary: Technical Abstract: Field trials are expensive and labor-intensive to carry out. Strategies to maximize data collection from these trials will improve research efficiencies. We have purchased a small unmanned aerial vehicle (AEV) to collect digital images from field plots. The AEV is remote-controlled and can be guided to a consistent height above the field plot, where it collects an image using a camera mounted on a gimbal. We are using it in several field trials. The first one is the National Verticillium Wilt trial, in which we have identical clones planted on Verticillium dahliae-infested and fumigated fields. Using a thermal infra-red camera, we are comparing leaf surface temperature in the two fields as a potential measure of disease resistance. We are also using a visible light camera to measure leaf color as disease progresses across the season. The digital image data will be compared with data collected on the ground. The second trial is one in which a set of clones is growing in one field under standard irrigation conditions and the other set is in a field with reduced irrigation. We are comparing leaf temperature in the two fields as a measure of drought tolerance. The third project is a yield trial consisting of 100 clones from a diploid segregating population. A recent study reported a negative correlation between canopy temperature and tuber yield, so we are testing that observation in our population. Finally, we are comparing visible and thermal infra-red images of early blight screening trials with data collected on the ground. AEV technology has the potential to provide extensive data sets at a low cost, with a small investment of time. |