Marine biologists from the ARC Centre of Excellence for Coral Reef Studies and the University of Queensland have increased the speed of underwater coral image analysis by a massive 3400%.
Dr Manuel Gonzalez-Rivero said almost 35 years of manual scientific analysis can now be completed within months.
“We’re also extracting and learning a lot more information from the images than first hoped and with a similar precision to human experts,” Dr Gonzalez-Rivero said.
“We have enhanced computer vision software, similar to facial recognition technology, to accelerate decisions about coral reef conservation efforts.”
This technology has brought major advances in medical science and his team, in partnership with UC San Diego, has spearheaded its application for coral reefs as part of the XL Catlin Seaview Survey.
The researchers are using automated image recognition software to rapidly and accurately examine coral reef communities across large areas to inform conservation decisions.
They have captured more than half a million images over three years across 1000 km in 21 countries and territories, creating a baseline analysis of coral reef health.
Dr Gonzalez-Rivero’s work is published in the journal Remote Sensing.
Contacts:
Dr Gonzalez-Rivero
E: m.gonzalezrivero@uq.edu.au
P: +61 7 3365 3452
Paper:
Manuel González-Rivero, Oscar Beijbom, Alberto Rodriguez-Ramirez, Tadzio Holtrop, Yeray González-Marrero, Anjani Ganase, Chris Roelfsema, Stuart Phinn and Ove Hoegh-Guldberg (2016) Scaling up Ecological Measurements of Coral Reefs Using Semi-Automated Field Image Collection and Analysis. 8 (1), 30; doi:10.3390/rs8010030.