“We’ve been using it for sheep, pigs and cows,” said Zhao Jinshi, who studied at Cornell University and founded Beijing Unitrace Tech, a company developing software for the agriculture industry.
“For pigs, it’s more difficult because pigs all look the same, but dairy cows are a bit special because they are black and white and have different shapes,” Zhao said as he checked on the technology installed in a pilot project here at a farm in Hebei province, outside Beijing.
China has led the world in developing facial recognition capabilities. There are almost 630 million facial recognition cameras in use in the country, for security purposes as well as for everyday conveniences like entering train stations and paying for goods in stores.
But authorities also use the technology for sinister means, such as monitoring political dissidents and ethnic minorities.
One Chinese AI company, Megvii, which has been blacklisted by the U.S. Commerce Department for alleged involvement in the Chinese government’s repression of Uighurs in Xinjiang, is applying its technology to a program to recognize dogs by their nose prints. Other tech companies around the world have had a go
at identifying chimpanzees, dolphins, horses and lions, with varying degrees of success.
What's the point?
Chinese entrepreneurs see an opportunity to apply this know-how to agriculture as farms become bigger and more commercialized, and as the rural population ages, limiting the number of people able to do manual work.
For humans, facial recognition works by measuring distances between features like eyebrows and lips. But for cows, the software detects patterns and shapes on the animals’ faces and hides. With 50 photographs from different angles, Zhao’s software can tell a Daisy from a Bessie.
Farmers load information such as health conditions, insemination dates and pregnancy test results into the system, which syncs up with cameras installed above troughs and milking stations. If everything works, farmers can amass valuable data without lifting a finger.
“We can monitor how long the cow is drinking for, how much it’s eating, how many times a day it visits the trough,” Zhao said as the cows walked in a row from their outdoor pen toward the milking shed.
Signs of illness or unusual behavior can be detected using artificial intelligence and treated quickly by a human, rather than relying on farmers to inspect the herd for potential problems.
“This system is very powerful and it will definitely make our work easier,” said He Ye, the manager of the farm in Hebei province. The farm has to buy the cameras, but Zhao’s company has been providing the technology for free while it irons out the wrinkles.
“When the weigh scale is installed, I will be able to monitor them in real time,” he said. If a cow shows symptoms of illness or any other problem, He gets an alert on his phone.
This kind of information for each animal used to be collected from electronic tags punched through their ears or worn around their ankles. The problem was that the cows were always trying to remove them — and often succeeding.
It’s a challenge to install the cameras in a farmyard, where there is water and mud everywhere. But even more challenging is the fact that cows are very curious, said Zhao.
“If you make a little change, they will notice it. If you put up a sign or change anything at all, the cow will notice,” he said. Then the cow will often try to nuzzle or lick or otherwise check out the new item.
This initiative fits nicely with the Hebei provincial government’s goal to double milk production within two years and improve safety.
“The milk industry has completely changed,” said He, the farmer. “The standards for milk production increased so much — it’s become much more rigorous, with requirements about sterilization and other sanitation regulations.”
Giving production a leg up
Long before the novel coronavirus — which scientists believe began in bats, then jumped through an intermediate host, probably pangolins, to humans working in an exotic meat market — China was known for low food standards and repeated scandals, such as melamine added to baby formula.
Millions of pigs were culled during a swine flu outbreak over the past year, and there have been several cases of bubonic plague, linked to marmots.
China is now paying more attention to food hygiene. With the spread of the coronavirus, China’s government has banned the trade and consumption of wildlife such as civet cats and bamboo rats.
“Intelligent farming” is beginning to change how the agricultural sector operates, and China’s technological giants are getting in on the action, focusing on pork, China’s favorite meat.
Alibaba, the tech behemoth, has been developing voice recognition technology for pigs to try to detect if the animals are in pain or trouble, while online retailer JD.com has been working on an AI-powered system to develop feeding plans for individual pigs.
This kind of technology could be useful on farms, said Gosia Zobel, a scientist with AgResearch, a New Zealand government institute.
“There are obvious constraints — like the technology working properly on a farm, appropriately validating what is being monitored, and data handling — but these challenges are worth tackling if they bring animal welfare benefits,” she said.
While farmers in other parts of the world are increasingly turning to technology to offset aging populations and labor shortages, China is mainly motivated to produce more domestically, said Salah Sukkarieh, a professor of field robotics at the University of Sydney.
“These technological advances are driving individual plants to get the most out of each square meter of land, to increase yields and reduce variability,” Sukkarieh said. “In China, this is mostly about food security and growing more on its national land.”
China feeds 22 percent of the world’s population with only 10 percent of the world's arable land. That creates extra incentive for China to improve food standards and production, including through the use of advanced technology.
If only those cows would leave the cameras alone and the pigs would develop more distinguishing features.
Wang Yuan contributed to this report.