We are working with Microsoft to enhance our Artificial Intelligence (AI) capabilities using high-resolution cameras to detect illegal fishing and help manage marine reserves.
Identifying illegal fishing behaviours
Signs of illegal fishing include the detonation of bombs being used for mass fish catch, diving activity for protected species, and fishing vessels that are setting fishing gear. However, monitoring illegal fishing activity can be extremely difficult.
Existing methods include ship tracking systems, fish catch data, and inspections at sea or in port. However, these approaches can be cost-prohibitive, particularly for developing countries. For example, countries such as Indonesia may have nearly a half of a million fishing vessels, so inspecting so many vessels is nearly impossible.
Cost-efficient data processing is needed to identify illegal fishing behaviours.
Next generation technology to manage illegal fishing
Focusing on users of marine reserves in Australia and fishing with explosives in Indonesia, we are combining robotic camera technology and underwater sound sensors called hydrophones to detect potential illegal activities.
High resolution cameras can capture data on the type of boat, boat features, boat travel speed, idling, and diving. Hydrophones can record sounds from vessel engines, air compressors, winches, or detonation of explosives from tens of kilometres away.M.
Together with Microsoft, we are developing new cost-effective data processing tools to turn this image and sound data into useable information.
Trained algorithms will be able to process this data to help detect patterns for IUU fishing. Further development will provide real time alerts directly of officials via satellite or mobile phone communications at very low cost. These alerts will equip authorities with timely information to investigate illegal fishing and support sustainable fisheries management.
Ensuring food security
Advances in digital machine learning is helping to expediate detection of illegal fishing.
We are improving the processing of large amounts of data using pattern recognition to detect and understand fishing patterns. Combining camera technology and audio recordings also provides the means to monitor illegal fishing during both day and night, and to integrate multiple types of data to build a more complete picture of unauthorized activities.
The application of low-cost innovative tools will assist fisheries management of marine resources, both in Australian and international waters, to protect the marine environment and help ensure food security and livelihoods for the world’s growing population.
Our aim is to further develop the technology to alert authorities in real time to support maritime response and decision-making.