Intel AI detects poachers in Africa
January 15, 2019
Intel will use artificial intelligence (AI) and IoT technology to detect poachers in real time at 100 reserves across Africa.
An elephant is killed every 15 minutes by a poacher, at a rate of approximately 35,000 elephants per year. In a decade, experts predict there won’t be any more elephants. Rhinos, gorillas, tigers and other large mammals are also in danger from poachers, as are giraffes, antelopes and wildebeest that are often caught in poachers’ snares.
Non-profit Resolve’s TrailGuard AI camera uses Intel-powered AI technology to detect poachers entering Africa’s wildlife reserves and alert park rangers in near real time so poachers can be stopped before killing endangered animals. TrailGuard AI builds on anti-poaching prototypes funded by the Leonardo DiCaprio Foundation and National Geographic Society.
The picture shows Intel’s Anna Bethke and Resolve’s Eric Dinerstein hide a TrailGuard AI anti-poaching camera in a tree during a recent demonstration in the mountains south of Monterey in California. The Intel Movidius Myriad 2 chip in the camera uses AI to identify potential poachers.
"By pairing AI technology with human decision-makers, we can solve some of our greatest challenges, including illegal poaching of endangered animals,” said Bethke. “With TrailGuard AI, Intel’s Movidius technology enables the camera to capture suspected poacher images and alerts park rangers, who will ultimately decide the most appropriate response."
TrailGuard AI uses Movidius vision processing units (VPUs) for image processing, running deep neural network algorithms for object detection and image classification inside the camera. If humans are detected among any of the motion-activated images captured by the camera, it triggers electronic alerts to park personnel so they can mobilise rangers before poachers can do harm.
"Reckless human activity is causing species loss and extinction on an unprecedented scale, with recent reports showing that as many as 60 per cent of all wildlife has been wiped out since 1970," said Justin Winters, executive director of the Leonardo DiCaprio Foundation. “If illegal poaching of wildlife continues at the current rate, elephants are just one of the large mammal species that will be completely erased in our lifetime. A commitment to protecting wildlife has been at the heart of LDF’s work from the beginning and we are excited to collaborate with Intel and Resolve on this breakthrough AI technology, which is set to be a game-changer for park rangers in the monitoring and management of endangered species around the world."
TrailGuard AI uses deep neural network algorithms that allow the device to recognise humans and vehicles with a high degree of accuracy. It builds on Resolve’s first-generation TrailGuard camera deployed in protected reserves that alerts rangers any time it detects motion. With the first-generation camera, rangers received many photos that they had to review manually to determine if there was a poaching threat or a false positive triggered by other motion. By adding an additional layer of AI into the system, TrailGuard AI intelligently knows when a potential poacher is present, allowing park rangers to intercept and apprehend rapidly.
The VPU, which delivers visual intelligence to the camera itself, resulting in several potential benefits:
- Limited false positives: Instead of alerting the rangers anytime there is motion in front of the camera, including from shifting cloud cover, birds and animals, TrailGuard AI only sends images to the rangers when a person or vehicle is detected. Limited false positives means rangers have more time to focus on their work, instead of spending their time looking through hundreds of false alerts each day.
- Long battery life: The VPU powers all TrailGuard AI’s processing needs – from wake-on-motion to image processing to AI inference to communications protocols – all while running at low power. It is designed to perform in the wild for up to 1.5 years without depleting the battery. This is an improvement over the original TrailGuard that has a separate computing unit requiring rangers to undertake the time-consuming and often dangerous task of field maintenance every four to six weeks. The long battery life also means less foot-traffic around the hidden cameras, which could alert poachers to their locations.
- Small size: Due to the size of the VPU, TrailGuard AI is about the size of a pencil and easier to hide and camouflage from poachers and wild animals.
In partnership with the National Geographic Society, Leonardo DiCaprio Foundation and others, TrailGuard AI will be deployed in 100 reserves in Africa throughout 2019, starting with Serengeti and Garamba, with plans to expand to south-east Asia and South America.
"Edge computing technology has the power to revolutionise the way we understand and protect our natural heritage," said Fabien Laurier, vice president of National Geographic Labs. "National Geographic is excited to work with Intel on TrailGuard AI and deploy these anti-poaching systems throughout Africa. This collaboration is critical to accelerating conservation and working towards our mission of achieving a planet in balance."