The application that assesses the quality of fish was created in Japan. Chefs will also have to photograph food

An application called Tuna Scope from Dentsu Inc uses machine learning algorithms based on thousands of tuna tail cross-sectional images that can tell a lot about fish structure.

In one photograph, the app can rate the fish on a five-point scale based on visual characteristics such as gloss and stratification of fat. For an experienced cook, these data say a lot about what kind of life a fish led, what it ate and how active it was – in such a way how the taste of the dish turns out.

Dentsu claims that its artificial intelligence captures “incomprehensible nuances of mastery in tuna research”, and during comparative tests with people who buy it, the application rated it to be more than four times out of five.

But land experts and fish producers are a little cautious about Tuna Scope’s ability to replace experts, especially those who buy meat for high-quality sushi and sashimi.

Keiko Yamamoto, a chef and land instructor from London, said that of course you can only evaluate tuna based on visual effects. According to Yamamoto, it seems possible to use artificial intelligence for basic quality assessments.

Right now, the Tuna Scope app is only used to evaluate fish for the Kura Sushi restaurant chain, which offers cheap sushi and uses other devices to save costs, such as robotic dishwashers.

“But although this kind of automation can work for such a large network as Kura Sushi, it will not meet the requirements of high-class cooks and sushi lovers”, – experts say.

Source: nv.ua