PoachNet integrates deep learning, GPS data, and elephant behaviour to help prevent poaching of Bornean elephants in the state of Sabah.
Asia’s smallest elephants can be saved from poaching using an artificial intelligence (AI) system developed by Cardiff University, researchers have said.
PoachNet integrates deep learning, GPS data, and elephant behaviour to help predict and prevent poaching of Bornean elephants in the eastern Malaysian state of Sabah.
“Elephant GPS data is analysed with a special type of AI — a sequential neural network – to predict their movements,” said Naeima Hamed, the lead researcher on the project. “These predictions are added to the knowledge graph in a meaningful way, then PoachNet uses a rule-based system to apply poaching rules and detect hidden patterns in the data.”
When tested, PoachNet was more accurate than other leading methods and performed better consistently.
Researchers claimed previous approaches to preventing elephant poaching have focused on specific aspects, such as social network analysis, multimedia data mining, or models based on ranger patrol data, but PoachNet approach incorporates an understanding of wildlife dynamics.
In the past, conservationists have been using automatic cameras to capture species, which could take months or even years to classify. However, the new system improves its predictive abilities by integrating advanced artificial intelligence. This allows it to detect illegal poaching activities more quickly by field intelligence.
PoachNet system can assist in informing strategies for tackling anti-poaching in Sabah, by helping with resource allocation based on its predictions. It can guide the deployment of motion-activated camera traps in areas most likely to have anticipated poaching crimes.
Bornean elephants are a subspecies of the Asian elephant. Scientists said due to habitat loss, human-elephant conflict, and poaching, less than 1,500 of them are left in the world. They have been listed in the International Union for Conservation of Nature’s (IUCN) Red List of Endangered Species.
“Despite global anti-poaching efforts, the illegal ivory trade continues to drive poaching, reducing the population to fewer than 1500,” said Benoit Goossens, a scientist at Cardiff University’s School of Biosciences.
The Cardiff research team plans to expand PoachNet by integrating other wildlife data sources like acoustic sensors to detect gunshots or vehicle noises as well as monitoring habitat changes and human activities through satellite imagery.
According to a paper published in the academic journal Science, around 28% of plant and animal species are facing the threat of extinction, highlighting the emergency for more wildlife monitoring by AI.
A conservation technology network named Wildlabs claimed AI is one of the three emerging technologies in the field of animal protection. They said AI can learn to identify rare species from thousands of photos, also locate animal calls accurately from hours of field recordings, greatly reducing the physical labor required to collect important conservation data.
“PoachNet equips biologists and conservationists with advanced tools for poaching predictions, offering a transformative approach to wildlife crime prevention,” said Dr Pablo Orozco Ter Wengel, Senior Lecturer in the School of Biosciences at Cardiff University.