Artificial intelligence for aquaculture has taken another step towards more widespread implementation, with a successful milestone reached by the German "Monitor Shrimp" project.
The research project, involving researchers from the Alfred Wegener Institute, together with industry partners Oceanloop and Sander Group, has reported up to 95% accuracy in detection and monitoring of shrimp health and wellbeing in land-based aquaculture operations, using computer vision and machine learning technologies.
The aim, the project partners say, is to support aquaculture companies by providing them with a better understanding of how their livestock are performing, as well as information on how to farm them ethically. Sustainability and animal welfare are increasingly important for Europe's aquaculture industry, and particularly within land-based shrimp farms, the project partners say.
One of the key participants in the project is German land-based shrimp farming pioneer Oceanloop, which currently operates two innovative land-based shrimp farms in Munich and Kiel.
Over the past two years, the researchers have developed a computer vision system that allows early detection and monitoring of shrimp growth, population size, mortality, and stress, reaching up to 95% accuracy.
This research was carried out real-time and under actual farming conditions, utilizing high stock densities and cloudy water. The first prototype was tested at Oceanloop's R&D farm in Kiel, Germany, in 2022.
Researchers installed a smartphone above the water's surface, programmed to photograph the shrimp once a minute, transferring the live data to a local server. Computer vision-based algorithms were then used to count the number of shrimp in each image and measure the lengths of individual animals, as well as to visually detect stress symptoms. The resulting data, the researchers say, can then be used to optimize growth and feeding models in real time.
"This function is used as an early warning system, but also to validate our stock density model. This bleeding-edge technology allows us to more precisely determine the density for each size group, which improves survival rates, growth, and feed efficiency," says Dr. Bert Wecker, Co-CEO of Oceanloop.
Closely monitoring mortality rates is crucial in aquaculture, according to Dr. Stephan Ende, AWI project coordinator.
"The mortality rates, which can range from 13 to 26 percent in salmon and can be up to 50 percent in shrimp, are substantially higher than with any other type of livestock. Online monitoring allows aquaculture companies to more quickly respond to signs of stress among their stock," he said.
The consortium also aims to "expand cooperation" with other interest groups "in order to spur on the digitalisation of shrimp farming", including ongoing collaboration with NeuroSYS, a team of experts in AI, machine learning, and digitalization.
Oceanloop's mission is to revolutionize the global shrimp industry using state-of-the-art aquaculture technology "made in Germany". Oceanloop's land-based shrimp farms enable high-quality, antibiotic-free, and sustainable saltwater shrimp farming where consumption takes place. Thanks to this approach, Oceanloop can offer considerably lower resource consumption and a far smaller ecological footprint than conventional farms. The Oceanloop Group currently operates two innovative land-based shrimp farms in Munich and Kiel. As the European technology leader for saltwater shrimp farming, Oceanloop's vision is to establish a global network of sustainable shrimp farms on land, contributing to the decarbonisation of the food and beverage sector.
