CRIMAC will improve and automate the interpretation of data and images from modern broadband acoustics on research vessels and fishing boats by using cruises and experimental field research, artificial intelligence, drones and inspection technology.
In addition to the main goal of more sustainable fishing and improved fisheries management, a general understanding of "acoustic dialects" could contribute to improved management in other uses of the ocean and coastal areas, such as within fish farming and energy production.
The Center has funding of about 28 million NOK a year for 8 years. It is a collaboration between research and industry.
CRIMAC = Center for Research-based Innovation in Marine Acoustic Abundance Estimation and Backscatter Classification.
Primary and secondary objectives
To advance the frontiers in fisheries acoustic methodology and associated optical methods, and to apply such methods to 1) surveys for marine organisms, 2) fisheries, 3) aquaculture and 4) the energy sector. This will be achieved via the following secondary objectives:
- To improve automatic interpretation of (wideband) fisheries acoustics.
- Aid the target classification of fish and zooplankton by experimental measurements and backscatter modelling.
- To collect reference data for machine learning projects, acoustics and optics on research and fishery vessels.
- Develop better verification methods using optical systems and dropped probes and mature working-drones.
- Work with user partners to apply developments in 1) to 4), in scientific surveys, for sizing and species classification in fisheries, for sizing, growth and behaviour in aquaculture, improved gas and bubble detection systems for the energy sector and presise bottom classification in the bottom mapping industry.
Fisheries acoustics, in sonar, sonars and fisheries sensors are used to measure and estimate the largest fish and krill populations in the world's oceans, and how they are distributed. A modern fishery without acoustic tools for detection, inspection and monitoring of the seabed, fish schools and the capture process itself is today unthinkable.
With newly developed broadband technology in the echo sounders and sonars, research and fishing vessels can now not only observe the echo strength and density of fish and other organisms under the vessel, but also observe the entire echo spectrum from the organisms.
For convenience, we prefer to define this as the "echo dialect" of the organisms. The echo dialect from a single herring is affected by body shape, swim bladder, body parts and its behaviour. It is very different from the "echo dialect" of mackerel.
In this research centre we will carry out systematic, experimental field research that can be used to understand and interpret the different echo dialects from fish, other marine organisms, gas emissions from the seafloor and various bottom substrates.
Machine learning techniques and AI
We will further expand existing multifrequency methods for automatic classification and target typing using machine learning techniques and artificial intelligence. A bat uses the same method and its life long experience to distinguish between the echo dialects from a fat fly and a thin fly, and a mosquito on the other hand. We intend to establish a knowledge base on echo dialects from fish, larvae, zooplankton, jellyfish and other objects, such as various bottom substrates and gas emissions.
This will improve the accuracy of existing acoustic interpreting methods and in the long term help the fishing skipper make good fishing decisions. Is it the right kind of fish she sees on the echo sounders, or is it a protected species? Furthermore, direct optical observations from the trawl with cameras and the use of active selection devices can reduce bycatch in the fisheries and allow fishers to target species or fish size more precisely.
Trawling with cameras and drones
For accurate verification of acoustic recordings, we will use the Scantrol DeepVision – a fish "photo booth" inside the trawl – equipped with active selection and open/close systems. Specific samples can thus be taken sequentially with trawl in deep water, for example inside schools of mesopelagic fish. Information from long trawl hauls with open cod end can also be used to give better observations of pelagic/bottom fish resources.
Miniaturized Broadband technology will be installed in probes, bottom rigs, observatories and surface and underwater unmanned vehicles (drones). CRIMAC will assess how such drones can support and improve scientific research surveys, but also how drones can be used in active fisheries.