NORA.startup webinar #29: Fish & AI
Interested to learn more about how artificial intelligence is being used in the Norwegian fishing industry?
Register to event by following the link below:
Registration is now closed
*This event will be streamed via zoom.
Artificial intelligence is a powerful tool that has immense potential within the fishing industry, considering the amount of data that is currently being collected. Consider for instance data on migration patterns that can be used to optimize catch, image processing that can be used to identify and classify fish species, satelite images that can be used to monitor illegal fishing activities, etc. Furthermore, by applying data data on fish and ocean, we can know more about environmental impact and status in specific areas, as well as existing opportunities and challenges amongst life underwater. Preserving and protecting life below water represents a vital block in the Sustainable Development Goals, namely in SDG14, which is also important in reaching SDG2, zero hunger, and SDG12, achieving responsible production and consumption.
During this NORA.startup event, we will explore exciting research conducted by researchers and innovative solutions provided by startups in the fishing industry in Norway. How can data for instance be used to reduce fish lice, predict migration patterns, monitor the well-being of fish, reduce carbon emissions or bycatch?
We will hear examples about how data can be utilized, the potential artificial intelligence has for the preservation and well-being of life underwater and future sustainable fishing practices, and a few other interesting examples of fish-related-innovation that might surprise you.
Startups, Projects & Speakers:
- Sergey Budaev & Jarl Giske, Professor, Department of Biological Sciences (BIO), University of Bergen
Fish wellbeing: a report from the inside
While it is hardly possible to see into the “mind” of the fish and infer their feelings, one can develop models that predict a range of welfare indicators from behavior and physiology. We have developed a causal agent-based simulation model of cognition, behavior, and stress that can be dynamically linked to environmental parameters in a facility. Model-based predictions of the fish internal states can provide an early warning system for upcoming stress and welfare challenges. The model also predicts fish behavior, potentially linking the observed behavioral patterns in the facility. This provides a quantitative approach for continuous monitoring, scenario modeling and predictive management of fish welfare state
- Kristoffer Løvall, Technical Manager, Scantrol Deep Vision AS
Scantrol Deep Vision, as part of the Scantrol Group, supplies innovative embedded systems for fish sampling - in the sea, onboard your vessel and in the lab. Our subsea camera unit is attached to the trawl, making it possible to identify and measure fish in the sea without bringing the catch onboard the vessel. This is an invaluable tool for marine researchers and a great technological leap towards more sustainable fisheries. We base our activities on the UN Sustainable Development Goals, and with the help of our partners, we develop resilient solutions for a greener future.
- Vaneeda Allken, Researcher, Institute of Marine Research
Vaneeda Allken is a machine learning researcher at the Institute of Marine Research. She has a Master’s in Physics from the University of Rennes 1 (France) and a PhD in Computational Geophysics from the University of Bergen in Norway. During a postdoctoral position at the University of Tübingen (Germany) in 2015, she started shifting towards more data-driven approaches. She used deep neural networks to detect signs of diabetic retinopathy from retinal images and developed a method for quantifying uncertainty in decisions made by the algorithm.
In 2017, Vaneeda started working at the Institute of Marine Research where her focus has been to develop practical machine-learning based solutions for problems in marine science. Her latest work aims at automating fish counting and the estimation of fish species distribution using images (Deep Vision) obtained from trawl surveys.
- John Costantino, Co-founder Manolinaqua
Manolin is a data company which analyzes data to assess risk profiles and benchmark performance across the aquaculture industry. We let farmers know when their fish health is in danger by constantly monitoring millions of industry data points and sending alerts whenever we see significant indicators of a decline in fish health.
Here’s how we do it:
We collected two decades of industry data and continue to monitor millions of production and environmental data points every day.
This is combined with a farmer’s production data in one easy platform, which automatically analyzes what’s happening in the water now and reports on the past.
Anonymized farm data is fed into our machine learning models, which send alerts when something is off and fish health is at risk, so farmers can act sooner.
This event is hosted in collaboration with
Read more about NORA.startup at www.nora.ai/nora-startup, and don't forget to apply for membership if you meet the criteria.