TrailGuard AI, an AI camera developed by non-profit RESOLVE is stopping poachers from entering Africa’s wildlife reverses and alerts the park ranges in near real-time.
The TrailGuard AI uses Intel Movidius Vision Processing Units (VPUs) to process images taken by motion-sensing cameras and then uses deep neural network algorithms to detect humans. If a human is found it triggers an alert to park personnel to apprehend and stop the poachers before the can kill the protected endangered animals.
RESOLVE, the non-profit behind TrailGuard, created and tested three protypes in Africa over last three years in collaboration with engineers from Intel Corporation.
The TrailGuard AI is built on anti-poaching prototypes funded by Leonardo DiCaprio and National Geographic Society.
Poaching is so widespread in Asia and Africa that it has driven several species to the brink of extinction. According to a report by National Geographic, Ploughshare Tortoises, that are hunted to supply as pets internationally due to their vivid gold and black shells, only 500 of them left.
Similarly, there are only 5,000 Black Rhinos left and are hunted for their horn
TrailGuard is a cheap and effective solution at stopping poachers
TrailGuard AI uses the Intel Movidius Myriad 2 VPU to deliver visual intelligence to the camera which reduces false-positives by effectively detecting people and differentiating them from other movements. TrailGuard AI also uses the Intel Movidius VPU for all its processing needs including wake-on-motion, image processing, and communication using very limited power reducing the upkeep requirements of the devices. The small size of the Intel Myriad VPU makes it easy to hide and camouflage from both poachers and wild animals.
The TrailGuard AI is being deployed at over 100 reserves across Africa in 2019, with the first ones being placed at Serengeti and Garamba. RESOLVE plans to expand to Southeast Asia and South America also.