High-throughput phenotyping has emerged as a promising area in plant, animal, and agricultural sciences that brings together researchers from life sciences, engineering, computer science, data science, mathematics, and other research fields to develop technologies for rapidly and accurately measuring phenotypes using robotics, imaging, and other tools. High-throughput phenotyping can be done at different scales, from cellular to ecological, typically using image-based approaches for data collection and analysis. The development of computer vision and machine learning approaches to extract biologically meaningful measurements from images, including physical, physiological, morphological, and qualitative properties of crops and livestock, is a major activity within the field. Phenotype datasets can be used for a variety of purposes, but in conjunction with large genomic datasets, are a powerful tool for linking phenotype to genotype, training genomic prediction models, and other approaches that integrate genetic, phenotypic, and environmental datasets.
We will introduce our efforts to develop PlantCV (
https://plantcv.danforthcenter.org/), an open-source platform for image-based plant phenotyping, and discuss opportunities for collaboration between the phenomics and database communities.
•• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• ••
Past webinars can be found
here.