ORCID

Abstract

The proliferation of accessible deep-water imaging platforms has resulted in the acquisition of vast amounts of image data, resulting in an analysis bottleneck. Object detection is now being applied to assist the image annotation process, with the potential to reduce analysis time. However, for object detectors to effectively tackle the scale of the challenge, models need to be generalisable to allow the transfer between imaging platforms and in space. This study trains YOLOv5 object detection models to identify six coral morphology groups using annotated imagery collected by ROV ISIS in the UK (JC136). Model performance was tested with independent datasets to inspect different aspects of transferability. Imagery collected on Tropic Seamount near the Canary Islands (JC142) with the same ROV (ISIS) was used to test spatial transferability. Imagery collected with ROV Holland I (SeaRover Project) from the Irish deep sea was used to test the transferability of models between ROVs. Model performance was moderate, recalling 60% of human annotations when evaluated against the validation dataset with varying performance across morphological groups (Recall = 44–69%). However, when tested using the independent datasets, model performance falls, recalling only 23% to 34% of human annotations across transfer scenarios. The results suggest that the model performance when transferred was poor, arising because of high shape variability within some morphological groups and poor taxonomic representation across datasets. We discuss how a coordinated community effort could improve model transferability and potentially address the analysis bottleneck.

Publication Date

2026-03-30

Publication Title

Ecological Informatics

Volume

95

ISSN

1574-9541

Acceptance Date

2026-03-28

Deposit Date

2026-04-20

Funding

We would like to thank Charlie Keeney, an undergraduate student at the University of Plymouth, for contributing to the image analysis. We would also like to thank the scientists, officers, and crews of all the research cruises that contributed to the collection of imagery data used in this study. The research cruise JC136 was funded by the UK Natural Environment Research Council, grant number NE/K011855/1 − DeepLinks project. The research cruise JC142 was funded by the UK Natural Environment Research Council (NERC) MarineE-Tech project, grant number NE/MO1151/1, awarded to Bramley Murton, National Oceanography Centre (NOC). Imagery data from the SeaRover programme acquired offshore Ireland during 2017, 2018 and 2019 were kindly made available by the Government of Ireland in support of this research. The Sensitive Ecosystem Assessment and ROV Exploration of Reef (SeaRover) was commissioned by the Marine Institute in partnership with the National Parks and Wildlife Service (NPWS) and funded by the European Maritime and Fisheries Fund (EMFF), Department of Agriculture, Food and the Marine (DAFM) and NPWS. The project was coordinated by the Department of Environment, Climate & Communications, funded INFOMAR programme team, with research support from the University of Galway, Plymouth University, and the Institute of Marine Research, Norway. INFOMAR is jointly managed by the Marine Institute & Geological Survey Ireland. This work was supported and funded by the Natural Environment Research Council, the ARIES Doctoral Training Partnership and the University of Plymouth; grant number NE/S007334/1.

Keywords

Cold-water coral, Computer vision, Deep learning, Model transfer, Object detection, YOLOv5

Download

Share

COinS