Deep Vision acoustic integration

Christian Michelsen Research has integrated support for Deep Vision images in their acoustic interpretation software LSSS. See video of the new functionality in use on a recent cruise conducted by the Institute of Marine Research (IMR), Norway.Images from the Deep Vision system are integrated with the acoustic data, providing species and size verification of targets seen acoustically. The images are placed at the proper geographic and depth position on the echogram, providing direct verification of the organisms present.– If a trawl sample contains multiple species or sizes, it is difficult to know what portion of the acoustic backscatter they correspond to, says Dr Shale Rosen, Research Scientist in trawl technology at IMR.– Now we can say definitively that this layer is mackerel, this is blue whiting, etc. In many instances we may be able to trawl with an open codend, collecting the species and size information we need simply from the images. And by identifying organisms before they reach the codend, species such as krill and small mesopelagic fishes that are too small to be retained in a fish trawl can still be accounted for.Rosen was one of the scientists who tested the new functionality during the recent REDUS (Reduced Uncertainty in Stock Assessments) cruise in the Norwegian Sea between Norway and Iceland. The cruise was also the first time the Deep Vision system had been deployed from a large modern commercial pelagic trawler/purse seine vessel.– Adding the Deep Vision data gives a much more complete description of what is going on than either the acoustics or trawl samples on their own. The Deep Vision data could often be used to identify the species and sizes comprising layers often seen in acoustic data. But equally importantly, they sometimes revealed a more complicated picture with multiple species present in a single image. For example, herring; blue whiting; mesopelagic fish and krill were all imaged together in the upper water column at night. Sometimes the Deep Vision images simplified the description of what was going on beneath the vessel, sometimes they indicated that the situation was more complicated than we assumed from the acoustics and trawl catches. But the important thing is that they give us a more accurate reflection of the true conditions, explains Rosen.– Images from Deep Vision revealed some surprising information to us, such as the discovery of cod more than 80 meters above the seabed. What we had initially interpreted as a dispersed layer of krill turned out to contain 50-80 cm cod which had likely swum up in the water column for a mid-night meal.Related:

Previous
Previous

Deep Vision research version ready for delivery

Next
Next

Students use Deep Vision to automatically recognize fish species