S³ 博客

关于进口、中国、亚洲和国际贸易的文章,帮助您做出更好的全球决策。

Chinese AI can now even recognise fish: artificial intelligence beneath the waters of Tibet

Facebook
Twitter
LinkedIn
WhatsApp
La IA china ya reconoce hasta a los peces: inteligencia artificial bajo las aguas del Tíbet

Artificial intelligence continues to find applications that, just a few years ago, would have seemed like science fiction. The latest comes from Tibet, where China is using AI to identify and monitor fish in one of its major rivers.

In this blog, we have often discussed the technological advances coming out of China. Humanoid robots, electric vehicles, automation, artificial intelligence, massive infrastructure projects… But even after so many examples, the Asian giant still manages to surprise us.

This time, quite literally, underwater.

China is using a system based on artificial intelligence and visual recognition to identify fish migrating along the Yarlung Tsangpo River, Tibet’s largest.

Or, to put it in slightly more striking terms: China now has ‘facial recognition for fish’, according to the South China Morning Post.

AI working underwater 24 hours a day

The technology has been installed at the Zangmu hydroelectric power station, whose dam stands 116 metres high.

To allow species to travel upstream, the facility has a fish passage system. And that is precisely where artificial intelligence comes into play.

Cameras and recognition algorithms make it possible to identify the species passing through the fish pass and automatically record their movements 24 hours a day.

The aim is to replace much of the manual work involved in observing and counting these movements, whilst at the same time obtaining much more detailed information on the behaviour of fish populations.

According to data compiled by the South China Morning Post, around 570,000 rare fish have used this passage to migrate upstream since the power station began operating in 2015.

Can AI really recognise a fish’s ‘face’?

The phrase is amusing, but behind it lies some rather serious technology.

Computer vision systems can learn to distinguish shapes, colours, proportions and other visual patterns found in animals. Using large quantities of images, algorithms are able to recognise certain characteristics and use them to identify or classify species.

This doesn’t mean that a fish will soon be unlocking a mobile phone with its face.

But the principle is inevitably reminiscent of the visual recognition systems we use with people: a camera captures an image, an algorithm analyses its characteristics, and the AI determines what it is seeing.

The big difference is the environment. Doing this underwater, with animals in motion and in natural conditions, adds a considerable technological challenge.

AI is moving beyond our screens

Perhaps this is the most interesting aspect of the news.

When we talk about artificial intelligence, it is easy to think immediately of conversational assistants, image generation or tools capable of writing text and analysing documents.

However, one of the great revolutions in AI may be taking place far from our screens.

Combined with cameras, sensors and computer vision systems, artificial intelligence is acquiring the ability to continuously observe the physical world, identify patterns and convert them into information.

It can analyse crops from the air, detect anomalies in infrastructure, recognise animals, monitor ecosystems or interpret vast quantities of images that would take a person weeks or months to review.

In this case, he is counting fish in a river in Tibet.

China is turning the region into a technological laboratory

This story also ties in with something we have seen time and again when discussing China’s technological advances: the enormous variety of settings in which the country is experimenting with new technologies.

Artificial intelligence is no longer the preserve of universities, research centres or large tech companies. It is finding its way into factories, cities, vehicles, farms, infrastructure and, now, conservation and environmental monitoring projects too.

And therein may lie one of the keys to the current technological race.

Because developing better algorithms is important. But finding thousands of places to put them to work is just as important.

The image might seem almost futuristic: a dam in Tibet, fish swimming through its structures and artificial intelligence silently observing every movement.

If we thought we’d already seen every possible application of AI in China, perhaps it’s best not to count our chickens before they’re hatched.

The next surprise could be anywhere.

Even underwater.

Facebook
Twitter
LinkedIn
WhatsApp

最新文章

公司新闻