The Department of Primary Industry and Resources (DPIR) has partnered with Microsoft in an award winning project recording, identifying and classifying fish on remote underwater videos using Artificial Intelligence.
The
Microsoft AI solution helps monitor and manage marine health by rapidly
analysing underwater video captured in Darwin Harbour.
The
platform automates the lengthy process of counting local fish stocks by
progressively learning to identify different varieties of fish species using
techniques similar to those used for facial recognition in social media.
DPIR
currently uses Baited Remote Underwater Video Stations (BRUVS) to
monitor fish species, however, there are difficulties in using BRUVS in Darwin,
where the tidal range is around eight metres.
DIPR fisheries
scientist, Dr Shane Penny, said the new AI platform was a great benefit to his
team as technical staff who would manually analysing BRUVS
footage are now redirected to other jobs, such as deploying and retrieving
BRUVS more frequently at more sites.
“This greater coverage then leads to more scientifically robust advice to fishery managers and the system is able to identify a fish in a video with 95 to 99 per cent accuracy.
“Two
particular species that have been of great interest to our team are golden
snapper and black jewfish as research has shown they had been overfished around
the greater Darwin area,” he said.
Part
of the management plan for those species was to put in place a number of
protection zones, with underwater video installed to monitor numbers in the
area.
Microsoft Australia’s National Technology Officer Lee Hickin said Cloud computing and AI were combining to support scientists to gain a deeper understanding of fish populations.
“Scientists
can now take the insights from the AI solution and focus on making informed
decisions that have significant environmental and economic impacts,” he said.
Displaying true innovation and collaboration, the partnership between DPIR Technology staff, DPIR Fisheries Researchers and Microsoft has resulted in a world leading Artificial Intelligence project putting Darwin firmly on the innovation map.
The Artificial Intelligence technique is known as Computer Vision and is described as the ability of a machine to recognise an object or shape from video or an image. It is a hugely complex undertaking and at the very forefront of Artificial Intelligence work around the world.
Media contact: Lucy Buhr 0419 106 309