AI saves endangered species by tracking gunshots

The AI-based tech could be used by conservationists all over the world. Source: kikatani (via Pixabay)
A new AI-based conservation technology may have a role in counteracting poachers by tracking gunshots across nature reserves and safaris. 

By Holly Giles | Deputy Editor 

Despite decades of education and efforts at prevention, the illegal wildlife trade is still estimated to be worth $20 billion annually, according to the WWF. It has led to dramatic population declines in species; elephants have been particularly targeted, as one of the “Big Five” of which 96 African Elephants are killed every day on average. Another example are tigers where in 1920 there were 100,000 of them and there are no only 3890. 

However, new AI technology may have a role in counteracting poachers by tracking gunshots across nature reserves and safaris. 

This technology relies on sensors which are able to analyse audio for gunshots up to 1km away. The sensors are then able to alert patrols.

This is a stark improvement on current options which use camera traps which are activated by movement. However, this is limited to close range and is much more expensive, meaning it is inaccessible for many conservation efforts. Conversely these AI-integrated audio sensors cost £50 each and are able to detect the noise from gunshots up to 1km away for 360 degree protection.

The technology has previously been used by police forces in the US and the West Midlands police force but has been removed from UK use due to technical difficulties; “One of the most challenging things with identifying gunshots is that the sound bounces off surface,” explained Professor Mark Plumbley, from the University of Surrey’s Centre for Vision, Speech and Signal Processing. This may make it perfect for the wide parks and plains present in conservation. Reflecting on this potential use Plumbley added: “I think it is an excellent application of this technology, to use it in conservation, where there are big open spaces and less air pollution. But there may still be vegetation of other things that get in the way”. 

The technology was tested through a collaboration between Google Cloud and the Zoological Society of London (ZSL) where they placed 69 audio recordings around the Dja Fuanal Reserve in Cameroon for a month. This amounted to 267 days of continuous sounds which could be analysed by the AI to detect gunshots and their respective location. 

It is thought this could pinpoint hotspots for poachers which the team could subvert through targeted patrols and a planned response. This was explained by Anthony Dancer, conservation technology lead at ZSL, who said: “Park staff can use [the information] to develop responses to those threats. Planning where to deploy partols in the areas and at the time of day where you most expect illegal activity”. 

Whilst this is a great advancement in the field, the team have further goals with the hope the sensors could one day identify humans talking, to indicate poachers were present before guns have been fired. This would reduce the amount of poaching able to take place and would put conservationists a step ahead; this increased knowledge for conservationists may help them win the fight against poachers. 

The head of customer engineering at Google, Omer Mahmood, added: “Animal poaching remains a global problem and with such catastrophic declines in some species, it’s an issue that cannot be ignored.”

“We’re committed to supporting ZSL and other conservation organisations with the best tools to tackle the current crisis,” he added.

This technology marks a major advance in the tools available to conservations in their fight against poachers. More technology gives them the upper hand and from this will help protect our animals. With current figures, these finds cannot come soon enough.

Science and Technology

Add Comment

Click here to post a comment

Your email address will not be published. Required fields are marked *