A new artificial intelligence program led by Google and Conservation International streamlines wildlife conservation monitoring to better protect animals.
Wildlife Insights uses A.I. and the Google Cloud, so conservationists — and anyone else in the world — can keep tabs on more than 4.5 million animals in the wild without actually disrupting their habitat.
“With this data, managers of protected areas or anti-poaching programs can gauge the health of specific species, and local governments can use data to inform policies and create conservation measures,” wrote Tanya Birch from Google and Jorge Ahumada from Conservation International in a blog post on Tuesday announcing the technology.
The program uses camera traps throughout the world to get up close and personal with reclusive or rare wildlife to better monitor them and their surroundings. The program also invites users to upload their camera trap images to grow the database further.
Other organizations involved in the project include the Smithsonian’s National Zoo and Conservation Biology Institute, World Wide Fund for Nature, Wildlife Conservation Society, and others.
The cameras can capture some pretty cute images, like close-ups of the noses of monkeys who growcurious about the camera or jaguar cubs rolling around in the grass. Since the data is open to the public, you can see what animals have been captured on camera in places ranging from the Central Suriname Nature Reserve in Suriname to the Lokaitra Forest in Madagascar to the Pasoh Forest Reserve in Malaysia.
Even though the data does show the general location of animals, Wildlife Insights does take steps to protect the actual location of certain species of animals.
“Wildlife Insights will obfuscate all geographic information made available to the public so that the exact location of a deployment cannot be determined from the data,” states Wildlife Insights’ website.
The ultimate goal for the data is to identify different species of animals faster. According to the blog, human experts can look through 300 to about 1,000 images per hour, but Google’s A.I. technology can analyze 3.6 million photos an hour and automatically classify an animal.
“Even though species identification can be a challenging task for A.I., across the 614 species that Google’s A.I. models have been trained on, species like jaguars, white-lipped peccaries and African elephants have between an 80 to 98.6 percent probability of being correctly predicted,” the blog explains.
Similar A.I. technology was created by a team from Duke University to identify up to 200 different species of birds just by looking at one photo. The tool was shown birds from ducks to hummingbirds and was able to pick out specific patterns that match a particular species of bird.
- AI turned Breaking Bad into an anime — and it’s terrifying
- Can A.I. beat human engineers at designing microchips? Google thinks so
- The BigSleep A.I. is like Google Image Search for pictures that don’t exist yet
- Google execs say we need a plan to stop A.I. algorithms from amplifying racism
- Sony wants to make cameras smarter with an onboard A.I. chip