Author Archives: Magnus Rogne Myklebost

AI-based net inspection tool

We are currently working on new machine vision based net inspection algorithms that combine both conventional and artificial intelligence (AI) based machine vision. The project is partially funded by FHF – Norwegian Seafood Research Fund and the tools developed with help the aquaculture industry reduce the risk of escaped salmon.

Automatic detection of damage to fishing nets is difficult to achieve with conventional machine vision algorithms. Weak contrast and poor visibility due to swirling debris and algae lead to a lot of false positive hole detections. The use of neural networks seems promising in dealing with these difficult conditions. 

Due to the large variations in water quality, net shape and foreign objects present at an aquaculture facility, we believe that using a combination of AI based and classical machine vision will give the best results. The system has to both be able to detect small holes before they represent a risk of escape and also not result in too many false positives.

The AI-based machine vision utilize Convolutional Neural Networks (CNN) and deep learning, where you present the training algorithm with tens of thousands of annotated images. Check out this nice article written by Henry Warren, that explains the CNN technology in an intuitive way.

The training takes up to 24 hours on a powerful server, and the result is a machine vision algorithm that is so efficient that it can run on a micro computer without access to the original dataset. The new machine vision has to be tested thoroughly in different conditions as there are a lot of pitfalls related to this technology. The prototype system will therefore first be used in combination with conventional inspection to prove it effectiveness.

New research equipment for remote areas

The decline in stocks of sea trout and wild salmon on the west coast of Norway has highlighted the need for a autonomous, non-destructive sampling method. For this reason, fish biologist at NORCE LFI are currently developing a trap with the overall goal of recording wild fish for research or conservation purposes.

Mohn Technology contribute with our high resolution FRS camera applying AI to detect passing fish. This new solar powered surveillance system was just deployed at Dale, Norway. The camera system is designed to use as little power and data traffic as possible, and the solar system was assembled and tested at Mohn Technology before deployment.

The system allows the researchers at NORCE LFI to get information about fish population and migrations directly to their computers from remote areas without access to electricity, and we hope more systems like this can aid research in this important field.

Automatic net inspection project grant from FHF

Mohn Technology is pleased to receive 5MNOK grant from FHF – the Norwegian Seafood Research Fund to develop a new autonomous net inspection drone for the aquaculture industry.

The drone will be a new tool to help fish farmers inspect their net pens more often and effective to a lower cost than diving and conventional ROV based inspection. An autonomous underwater vehicle (AUV) is a very complex product, that will demand a lot from our developer team. Mechanical and cybernetics has to work hand in hand in order for the AUV to work efficiently and effective under difficult conditions.

The AUV will be battery powered without umbilical, to reduce the risk of entanglement in existing equipment in the facility. Navigation will be based on a mixture of machine vision, compass and IMUs (Inertial Measurement Unit). The customer will gain access to reports and status via a web portal.

We are really looking forward to start on this project, as it is both very interesting and fits our company profile well!

Many thanks to FHF for the trust and support!

Master thesis with UiB

Mohn Technology has in cooperation with UiB Department of Informatics and masters student Oda Inanna Klemetsdal Stene started a project regarding AI based image segmentation. Oda will work on creating a new machine vision tool for the fisheries industry that will increase profits and reduce waste. We are looking forward to the collaboration.

If you wish to write bachelor or masters thesis with us, please contact magnus (at) mohntechnology.no

New FRS camera in the water

NORCE Research has installed a new FRS camera in the Bolstad River. It was important to get the system up and running before the spring flooding due to snow melting.

The system was installed by NORCE field biologists, who bolted the durable stainless steel frame to a large boulder . The boulder was then moved to deeper waters. The frame / bracket is designed to withstand heavy impacts by objects that float down the river.

Pelagic fish project with IMR

Mohn Technology and the Norwegian Institute of Marine Research (IMR) are working together to create new tools to improve the accuracy of purse seine fisheries. The project consists of many technological advancements, and Mohn Technology will contribute with our machine vision and remote controlled vehicle knowledge.

The results will help the fishing industry increase their pre-catch knowledge about the fish and improve profits. The tools can also be used by researchers who wish to learn more about the pelagic fish population.

Two FRS pilot projects live

In a cooperation project with NORCE Research, we have delivered our two first pilot systems of the FRS cameras to BKK. BKK, who is interested in the local fish population and migration patterns, has installed the systems at two suitable fish ladders. We are looking forewards to working together with both BKK and NORCE to further develop the system and implement customer ideas and requirements.

AI based fish detection machine vision

Mohn Technology is continuously working on testing and improving our machine vision algorithms for our different camera systems. After we completed a prototype test of our FRS camera (Fish Research System) this winter with NORCE LFI at Byglandsfjord, we got new footage to test.

Machine vision is one of the most important aspects of the FRS camera, and high precision algorithms help us deliver the best possible results. The FRS allows the user to efficiently monitor underwater life with greatly reduced manual data processing efforts.

This version was trained on a small data set, but the results were pretty impressive. The machine vision algorithm detects fish that are hard to see manually (especially those inside the net).

The system provides both still images and video, which makes it easy to verify and classify the fish.

Theis particular setup, with a fishing net funneling fish past the FRS camera, is used by NORCE to research fish population in lakes and fjords. It is one of many ways to use the FRS camera to gather data about marine life.