These researchers are using AI drones to keep track safer wildlife

These researchers are using AI drones to keep track safer wildlife

In the late ’90s, environmentalists wildlife Zoe Jewell and Sky Alibhai struggled with a disturbing realization. The couple had studied the black rhino populations in Zimbabwe, with tranquilizer darts and spent a good portion of their time, the animals resumed and attach radio collars around their necks. But after years of work, researchers have realized it was a big problem: your technique, commonly used by all types of wildlife scientists seemed to have rhinos female to cause fewer offspring. The researchers published their findings in 2001, a dispute in the World Conservation ignite. The problem, says Duke University conservation ecology professor Stuart Pimm, is that it is “clicked”, is extremely stressful for the animals. “If you were walking through your neighborhood and suddenly a couple of strange people came to the store … and you shot him in the ass with a dart, and I woke up with something around my neck, I think it’ll be pretty bad shape would “, He says. But Jewell and Alibhai had an idea. While the works next to the Shona tribe of Zimbabwe, have seen how the trackers indigenous capable of a huge amount of information about the footprints of wildlife animals, including the weight, sex and species, without which all originate from somewhere near the animals themselves. “We went to the local game scouts, who often were expert trackers, and they would laugh with us often, as we refer to these signals from the collar being heard,” said Jewell. “Told us, ‘all you have to do,” it might be possible to look at the ground, you are asked to apply these techniques conservation efforts time-tested Yes, it turns out.?. In 2004, the track two wild, the wildlife researchers to pursue a fingerprint identification technique (FIT) and special software developed populations nonprofit organization founded. Wild Track can identify with 95% accuracy individual animals, it is said, using all without getting up close and camera traps. In addition, he says Jewell Wild Track sprayed one of the central challenges in any conservation effort to help solve: find out how many animals are in nature, and where exactly they are. While the process was inspired by indigenous trackers, he says Jewell Wild Track technology continues to be, compared to their capabilities are limited. “And ‘humbling in a way to see what these [Tracker] experts can do in the area,” he says. “I can only a little ‘, to emulate that.” Wild Track Process is tens of conservation efforts around the world, from tracking rhinos in Namibia otters in Portugal for the Tigers in Nepal. But to identify in China, where the Panda researchers previously little choice, the length of the bamboo pieces of undigested in the feces to measure individuals and keep track of FIT protocol that is faster and more accurate, was a major step forward . “This is extraordinary, the exciting work,” says Pimm. “And ‘technically very clever, and feeds directly into good natural history.” Search With hundreds of hours every week fingerprint images efforts around the world by tracking, the researchers are in the wild track to new advances in the computer industry before any artificial intelligence (aI) to speed up their work. Jewell and Alibhai have with SAS software provider since the late nineties. In 2018, they worked to create a machine learning algorithm to automatically identify fingerprints or markers in the fingerprint images, a task for researchers to conduct found by hand. early to recognize a system of artificial intelligence team, the various types of big cats trained straight looking prints. “After a while ‘you get to understand the difference between a puma and a snow leopard, or one of the really big cats,” says Craig Mann, a software developer for the wild track project at SAS. Although the AI ​​software does not do all the human research work, FIT process about the facts in terms of time five times faster. AI advances will help conservation efforts in the laboratory not only used – artificial intelligence researchers also wild track will improve the collection of about their technology presence. to continue cooperation with European companies senseFly drone, develop a fingerprint AI-powered drones and collect information about the animals from the air. In August, the researchers tested used a drone usually for the extraction, detection and Agriculture, to scan a large area and to find traces of people, dogs and horses. Their hope is to develop a drone, the resolution scan fingerprints, as well as follow animal tracks through fields or deep snow. Some of the wild interest in working width has attracted outside the conservation community. The group has received financial support from the US Army, projects often funded, where we see a potential long-term benefits. Some of these interests came from his army conservation work that will assist Park Rangers includes preventing poaching and illegal trade of elephant tusks and rhinoceros horns, activities doses fund extremist groups. But it is also interesting to follow similar techniques in dealing with people. “We see this as a long-term investment,” says Stephen Lee, a senior scientist at the US Army Research Laboratory. “[We] have been developed techniques can of course investigate when individuals applied to identify specifically.” For Jewell, it is a happy coincidence that with which to align the goals of the US Army of wild track. Dice stop the trade of body parts such as Rhino horns of animals in danger of extinction it is advantageous, both for the conservation and regional stability. “We are very encouraged that the US Army Research Office is working with us to ensure the conservation of species in danger of extinction and support glad that we have a common mission,” he said via e-mail. The military research may eventually pay dividends in the defense sector, but is also a powerful reminder of the potential of all technologies, including those developed in different ways to order a pleasant as species protection, it is used as originally intended. But faced with a number of environmental threats – poaching, habitat loss, climate change – can not be an alternative, but to prevent any technological progress and financial options available to use the extinction of vital species. And if this means funded rhino monitoring of the military, AI-powered drones tracking footprint glide high above the savannah, then so be it. “Noninvasive, I think it’s much more than an idea that the time has come,” says Pimm. “It will be the wave of the future.”
Picture copyright by Rocket Wolfgang Kaehler-Light via Getty Images