Subtidal Seagrass Detector: Development and preliminary validation

Langlois, Lucas A., Collier, Catherine J., and McKenzie, Len J. (2022) Subtidal Seagrass Detector: Development and preliminary validation. Report. TropWATER, James Cook University, Cairns, QLD, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website:


This report presents the development and evaluation of a Subtidal Seagrass Detector (the Detector). Deep learning models were used to detect most forms of seagrass occurring in the northeast Australian seascape from underwater images and classify them based on how much seagrass was present. Images were collected by scientists and trained citizen scientists undertaking routine monitoring using drop-cameras mounted over a 50 x 50 cm quadrat. The Detector is composed of three separate models able to perform the specific tasks of: detecting the presence of seagrass (Model #1); classify the seagrass present into three broad cover classes (low, medium, high) (Model #2); and classify the substrate or image complexity (simple of complex) (Model #3). We were able to successfully train the three models to achieve high level accuracies with 97%, 80.7% and 97.9%, respectively. With the ability to further refine and train these models with newly acquired images from different location and from different sources (e.g. ROV), we are confident that our ability to detect seagrass will improve over time. With this tool we will be able rapidly assess a large number of images collected by various contributors, such as citizen scientists, QPWS Rangers and Indigenous rangers that frequently access the Reef and seagrass habitats of northern Australia. This would provide invaluable insights about the extent and condition of subtidal seagrass in currently data-poor areas.

Item ID: 80573
Item Type: Report (Report)
Keywords: seagrass, deep learning models, monitoring
Related URLs:
Copyright Information: © James Cook University, 2022
Date Deposited: 25 Sep 2023 23:04
FoR Codes: 41 ENVIRONMENTAL SCIENCES > 4199 Other environmental sciences > 419999 Other environmental sciences not elsewhere classified @ 100%
SEO Codes: 18 ENVIRONMENTAL MANAGEMENT > 1899 Other environmental management > 189999 Other environmental management not elsewhere classified @ 100%
Downloads: Total: 1
Last 12 Months: 1
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page