In Japan, caviar, which is prepared from salted sturgeon eggs (or roe), is a pricey delicacy. Due to overfishing and pollution, wild sturgeon is seriously threatened, making caviar a luxury good. The harder it is to find caviar, the less sturgeons there are to produce roe.
Sturgeon farming in Japan hasn’t grown much since it requires a lot of time and work; for example, the fish must be raised for at least six to seven years before they reach sexual maturity.
Aquaculture, however, may be a solution to assuring consistent caviar supply. Complicating matters is the fact that breeding workers must have specific knowledge and conduct long-term monitoring because male and female sturgeons are difficult to identify for two to three years after birth.
Overfeeding or environmental changes, such as changes in water temperature, can cause losses in inventories.
According to Japanese telecoms and internet company Softbank Corp., artificial intelligence (AI) and Internet of Things (IoT) technologies may offer a solution to these problems. Since 2020, the business has collaborated with the Hokkaido University Faculty of Fisheries Sciences and the town of Bifuka, located in Hokkaido, to develop low-cost, effective sturgeon farming techniques utilizing AI and IoT.
They are used to automatically detect, count, and track sturgeons, as well as to identify the individual swimming patterns of fish and imitate schooling behavior by analyzing visual data. In order to significantly increase the production and quality of sturgeon, the parties inked a three-way industry, academic, and government partnership agreement in March.
Yasuaki Takagi, head of the Faculty of Fisheries Sciences at Hokkaido University, stated to the Advocate that one of the faculty’s objectives is to revive the fisheries in Hokkaido.
Since the 1980s, we have been promoting the area through sturgeon farming in Bifuka town, one of the local municipalities in the prefecture, with whom we have collaborated for many years. The town’s sturgeon aquaculture is expanding right now, and with it comes a growing need to use technology to improve agricultural practices.
Must read:China warns of artificial intelligence risks, calls for beefed-up national security measures
Yuko Ishiwaka of Softbank stated, “Our objective is to develop a new, superior strain of sturgeon that has never been farmed.”
“We will collect the required data using our own technology, the high-level rearing methods being used in Bifuka Town, and the technology and sturgeon knowledge of Hokkaido University.”
The partnership will make use of Softbank’s image analysis and machine learning technology to recognize specific sturgeons being raised at the aquaculture plant in Bifuka Town, analyze their behavior, and create simulations.
The ecology, anatomy, and musculoskeletal structure of the sturgeon will be studied by Hokkaido University, while Bifuka Town will oversee putting farming techniques based on Softbank’s technology and the aquaculture and biology knowledge of Hokkaido University into practice.
At Hokkaido University, work started in 2020 with the dissection of sturgeon. The bones and muscles were then meticulously sketched. After that, Softbank spent three years analyzing the connections between the sturgeon’s bones and muscles to create a 3DCG (three-dimensional computer graphics) model using anatomical data from Hokkaido University.
Since then, Softbank has created a 3DCG simulation model that includes swarming, schooling, and milling as swimming behaviors.
Sturgeon cultivation is quite difficult, according to Takagi. First, additional study is required to better understand the reproductive physiology of sturgeons to increase the stability of egg and sperm production.
Additionally, we require a consensus definition of a “good sturgeon” and careful parent fish breeding. We also need farms that can generate a consistent supply of seedlings because sturgeon take a very long time to mature compared to other species and growing them in research facilities in Japan is challenging.
Takagi claimed that while it is possible to bring in high-quality seedlings from outside of Japan, doing so entails a danger of bringing in infections. Additionally, he stated that sturgeon disease prevention measures must be put in place: Since there hasn’t been much research in this area in Japan, “it will be impossible to act when disease occurs” without professionals.
“We will be able to simulate other aspects, such as spawning behavior in tens or hundreds of generations of sturgeon,” said Ishiwaka, “if we can study the ecology of sturgeon even further and simulate not just their musculoskeletal growth but also other areas like organ development or reproduction.”
Eventually, we could imitate the key aspects of a sturgeon farm’s rearing procedures and learn more effective farming methods.
It is crucial to be able to model the processes involved in raising, according to Takagi. This approach may eventually result in a regular supply of farmed sturgeon, which could lead to new technical developments in aquaculture and better farming techniques for other species.
Additionally, simulations can be used for early planning, to gather and show specific data on individuals who have been raised, and as a tool for making decisions that will improve farm management.
It is obvious that we cannot rely solely on fisheries as a supply of marine protein as the world’s population rises, according to Takagi. “Aquaculture is receiving a lot of attention, and our work is crucial to its development.
A large-scale investment is also needed to develop an industry like aquaculture, which might be quite risky for some nations. To lessen that risk, we think it is crucial to be able to imitate rearing techniques using AI and IoT.
Must read:Hollywood stunt performers face danger from Artificial Intelligence