AI Cameras Make Huge Leap in Protecting the Rarest Animals on Earth

Pangolin in the wild

Artificial intelligence-powered cameras are making it possible to track endangered species in real time, allowing conservationists to respond quickly and protect these rare creatures from threats like poaching or fires.

The screen capture above and the video below show one of the first times that a pangolin — a critically endangered species — has been detected in real-time using artificial intelligence (AI). It was spotted by United Kingdom-based nonprofit Conservation AI using a camera powered with NVIDIA technology.

Conservation AI has spent the last year and a half deploying more than 70 AI-powered cameras around the world that are designed to help conservationists preserve animals thanks to real-time detection using deep learning models trained with what is known as transfer learning — the ability to take a neural network that was designed for one purpose and apply it to another.

The nonprofit uses the NVIDIA Jetson edge AI platform and NVIDIA Triton Interference Server — neither of which were originally designed for conservation purposes — to analyze footage, identify species of interest, and alert conservationists of potential threats via email in a matter of just four seconds.

In addition to real-time reaction, the AI system can also model trends in biodiversity and habitat health using a large database of photos and other metadata that would normally take a person several years to analyze manually.

“A typical camera trap study can take three years to analyze, so by the time you get the insights, it’s too late to do anything about the threat to those species,” Paul Fergus, a professor of machine learning at Liverpool John Moores University, explains.

“Conservation AI can analyze the same amount of data and send results to conservation teams so that interventions can happen in real-time, all enabled by NVIDIA technology.”

The Conservation AI platform has been deployed by Chester Zoo — a conservation society based in the UK — to detect poachers in real-time, including those hunting pangolins in Uganda.

“Since many endangered species, like pangolins, are so elusive, obtaining enough imagery of them to train AI models can be difficult,” NVIDIA explains. “So, the Conservation AI team is working with NVIDIA to explore the use of synthetic data for model training. The platform is also deployed at a game reserve in Limpopo, South Africa, where the AI keeps an eye on wildlife in the region, including black and white rhinos.”

The organization’s long-term goal is to expand the amount of data available to the aI even further by integrating satellite imagery and audio analysis such as an animal’s cries or the sound of a forest fire. The more data conservationists have and the faster they can analyze that information, the easier it is for them to pursue the greater goal of boosting their conservation efforts and fighting poachers.

Image credits: Pangolin footage courtesy of Chester Zoo.