Nondestructive automated workflow for analyzing diverse leaf morphologies using computed tomography

Abstract

Automated plant analysis methods can inform basic and applied plant research. Computed tomography (CT) has been used to image plants in canopies and individual plants; however this method is underutilized for dynamic analysis of plant growth. In this work we present a workflow and associated algorithm to nondestructively extract leaf area from 120 individual CT scans for plants with flat [soybean, Glycine max (L.) Merr.], textured (tomato, Solanum lycopersicum L.), and grassy (wheat, Triticum aestivum L.) leaves. Under low water conditions, we see significant changes in leaf area depending on leaf type. This work enables future automated phenotyping using CT scanning of whole plants.