Variation in forest root image annotation by experts, novices, and AI

Grace Handy, Imogen Carter, A. Rob Mackenzie, Adriane Esquivel-Muelbert, Abraham George Smith, Daniela Yaffar, Joanne Childs, Marie Arnaud

Research output: Contribution to journalArticlepeer-review

Abstract

Background: The manual study of root dynamics using images requires huge investments of time and resources and is prone to previously poorly quantified annotator bias. Artificial intelligence (AI) image-processing tools have been successful in overcoming limitations of manual annotation in homogeneous soils, but their efficiency and accuracy is yet to be widely tested on less homogenous, non-agricultural soil profiles, e.g., that of forests, from which data on root dynamics are key to understanding the carbon cycle. Here, we quantify variance in root length measured by human annotators with varying experience levels. We evaluate the application of a convolutional neural network (CNN) model, trained on a software accessible to researchers without a machine learning background, on a heterogeneous minirhizotron image dataset taken in a multispecies, mature, deciduous temperate forest. Results: Less experienced annotators consistently identified more root length than experienced annotators. Root length annotation also varied between experienced annotators. The CNN root length results were neither precise nor accurate, taking ~ 10% of the time but significantly overestimating root length compared to expert manual annotation (p = 0.01). The CNN net root length change results were closer to manual (p = 0.08) but there remained substantial variation. Conclusions: Manual root length annotation is contingent on the individual annotator. The only accessible CNN model cannot yet produce root data of sufficient accuracy and precision for ecological applications when applied to a complex, heterogeneous forest image dataset. A continuing evaluation and development of accessible CNNs for natural ecosystems is required.

Original languageEnglish
Article number154
JournalPlant Methods
Volume20
Issue number1
DOIs
StatePublished - Dec 2024

Funding

The work presented in this article is supported by the WoodLand Trust small research grant 1975983: \u201CGetting to the roots of the forest response to elevated CO2\u201D (2022) awarded to M Arnaud, A Esquivel Muelbert and Rob Mackenzie. Images were sourced from the Birmingham Institute of Forest Research Free Air CO2 Enrichment (BIFoR FACE) facility, funded by The JABBS foundation, The University of Birmingham, and The John Horseman Trust. Grace Handy was supported by NERC CENTA2 grant NE/S007350/1. Marie Arnaud was supported by HORIZON-MSCA\u20102021\u2010PF\u201001 grant funding to Sorbonne University, grant no. 10106240. ARMK gratefully acknowledges support of the UK Natural Environmental Research Council through grant NE/S015833/1 (QUINTUS). The work by Abraham Goerge Smith on this article is supported by Novo Nordisk Foundation grant NNF22OC0080177. Daniela Yaffar thanks the Next Generation Ecosystem Experiments-Tropics, funded by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research, and the Oak Ridge National Laboratory managed by UT-Battelle, LLC, for the DOE under contract DE-AC05-1008 00OR22725.

Keywords

  • Forests
  • Image analysis
  • Machine learning
  • Minirhizotron
  • Root annotation
  • Roots

Fingerprint

Dive into the research topics of 'Variation in forest root image annotation by experts, novices, and AI'. Together they form a unique fingerprint.

Cite this