(714.13 MB)
Download file


Download (714.13 MB)
posted on 04.08.2021, 17:51 by Liam LachsLiam Lachs, Fiona Chong, Maria Beger, Holly EastHolly East, James GuestJames Guest, Brigitte Sommer


The goal of SizeExtractR is to facilitate scientific projects that have research questions relating to size, from pure biology to ecology on individual to landscape scales, by providing a protocol with a user-friendly set of interactive tools for image analysis (ImageJ-macros) and database formation and quality checking (R-package).

Uses include demographics, population dynamics, and size as a proxy of energy investment.

Image Analysis

The ImageJ tools and protocol serve to optimise the slow task of manually outlining objects from sets of images. A scale is required within each image. These tools can save considerable time, and are a most-suited to image analysis applications where machine-learning or other automated outlining classifiers are innefective or out of scope.

R package

The R tools are used to synthesis all data into one master dataset, with all variables of interest included. Quality control tools are integrated into the database formation. A ggplot2-based plotting function to visualise size-frequency distribution is also provided. R-package can be accessed through 10.5281/zenodo.5997934.

Size Metrics

Measures of size that are captured using SizeExtractR tools include:

  • area
  • minimum and maximum diameter
  • estimated spherical volume
  • perimeter length

Other Metrics

Users can also record additional categorical variables of interest using a simple labeling system. For instance, the tools were developed on a coral bleaching demography data set, so coral species, bleaching status, partial mortality, and parasitism were all recorded as categorical annotations relating to individual outlined objects (corals).

Contents of Data Repository

Relevant data and code for:

  • SizeExtractR ImageJ macros
    (on Github: SizeExtractR/inst/
  • SizeExtractR R package
  • Timed image analysis


Natural Environment Research Council’s ONE Planet Doctoral Training Partnership (NE/S007512/1) to LL

Natural Environment Research Council’s Panorama Doctoral Training Partnership (NE/S007458/1) to FC

European Research Council Horizon 2020 project CORALASSIST (725848) to JRG

Australian Research Council Centre of Excellence for Environmental Decisions (CE110001014) to MB

EU Marie Skłodowska-Curie Fellowship (TRIM-DLV-747102) to MB