Phenotyping system for precise monitoring of potato crops during growth

Abstract

A phenotyping system was developed for monitoring the growth of 3 Japanese and 5 European varieties of potatoes. From the UAV images taken weekly, plant height, volume and vegetation indices were estimated from the DSM and the reflectance maps. UAV-estimated volume and manually measured plant height had a high correlation with a coefficient of determination R2 of 0.70. UAV estimated NDVI was highly correlated with the manually measured NDVI (correlation coefficient of R ＝0.97). European varieties such as Euroviva had a higher volume compared to the Japanese varieties such as Irish cobbler and Toyoshiro. Volume and LCI were found to be highly correlated with the yield (R2 of 0.99 and 0.96) thus ideal parameters for yield prediction.