Machine learning techniques to characterize functional traits of plankton from image data

Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab-based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have potential well beyond examining the abundances of different taxa; the individual images themselves contain a wealth of information on functional traits. Here, we outline traits that could be measured from image data, suggest machine learning and computer vision approaches to extract functional trait information from the images, and discuss promising avenues for novel studies. The approaches we discuss are data agnostic and are broadly applicable to imagery of other aquatic or terrestrial organisms.

Automated summary

Plankton imaging systems, supported by automated classification and analysis, have enhanced the ability of ecologists to observe aquatic ecosystems. These systems enable the collection of imaging data at unprecedented levels of spatial and temporal resolution, allowing for reliable tracking of plankton populations. Additionally, the individual images themselves contain valuable information on functional traits, which can be extracted using machine learning and computer vision techniques for further analysis and novel studies.