Phenotyping Morphological Traits of Blueberry Bushes using UAS

Abstract. Phenotyping morphological traits of blueberry bushes in the field is important for selecting genotypes that are easily harvested by machine and have high yields. In addition, the rate at which a bush grows is indicative of its vigor, and is therefore of interest to breeders and growers. Morphological data can also be used to assess the effects crop treatments such as plant growth regulators, fertilizers, and environmental conditions. This paper investigates the feasibility and accuracy of an inexpensive unmanned aerial system in determining morphological characteristics of blueberry bushes. Color images collected by a quadcopter are processed into 3-D point clouds via structure from motion algorithms. Bush height, extents, canopy area, volume in addition to crown diameter and width are derived and referenced to ground truth. Twelve bushes were imaged by a quadcopter in 8 minutes, and height and width dimensions achieved a mean absolute error of 8.73cm. Strong correlation was found between manual and image derived bush volumes and their traditional growth indices. The methodology developed from this study is suitable for high throughput phenotyping for blueberry breeders, and producers, and provides a more intuitive metric for bush evaluation than current growth indices.