The Cacophony Project, a New Zealand based organisation, is using the latest technology to lure, identify and eliminate invasive predators. This involved heat cameras, AI identification, sound lures and novel tools to control invasive species.
Grant Ryan, The Cacophony Project
The conservation challenge
New Zealand is one of the worst places in the world for the number of species endangered or still in decline. This is because New Zealand native fauna did not evolve with predators and introduced species can decimate the number and diversity of local birds and insects.
The first part of the solution is a heat camera that can identify all introduced predators in the field. The most common tools used at the moment are tracking tunnels and chew cards. Even off the shelf tracking cameras are designed for pigs and deer rather than smaller faster mammals.
To make cameras labour-efficient and accurate, the team at the Cacophony project applies AI to identify animals to the species level (i.e. rats, stoats, possums, cats, birds). The team has developed the most sensitive tool to monitor introduced predators in New Zealand. Their cameras have over 95% accuracy so only 5% of footage need to be looked at manually.
The use of machine learning algorithms is a relatively easy process compared to the large amounts of data that the team needs to manage. A crucial step is to ensure data is properly tagged so that the machine learning algorithms have high quality to “learn” from. The heat camera is currently more sensitive and cost effective to monitor predator numbers than any other tool.
Using artificial intelligence algorithms, the team is able to rapidly test digital lures (social sounds) and apply big data tools to test potential lures in the field.