Space satellites use machine learning to count elephants on Earth

Vulnerable elephant populations are now being tracked from space using Earth-observation satellites and a type of artificial intelligence (AI) called machine learning.
As part of an international project, researchers are using satellite images processed with computer algorithms, which are trained with more than 1,000 images of elephants to help spot the creatures. 

With machine learning, the algorithms can count elephants even on ‘complex geographical landscapes’, such as those dotted with trees and shrubs.   

Researchers say this method is a promising new tool for surveying endangered wildlife and can detect animals with the same accuracy as humans.

The technique allows vast areas of land to be scanned in minutes, offering an alternative to human observers counting individual animals from low-flying planes.  

With effective monitoring, researchers hope to help stabilise and boost the population of African elephants (Loxodonta africana), which has plummeted over the last century due to poaching, retaliatory killing for crop raiding and habitat fragmentation. 

With somewhere between only 40,000 and 50,000 elephants left in the wild, the species is listed as ‘vulnerable’ on the IUCN Red List of Threatened Species.   

‘Accurate monitoring is essential if we’re to save the species,’ said study author and computer scientist Dr Olga Isupova at the University of Bath, who created the algorithm that enabled the detection process.

‘We need to know where the animals are and how many there are.’ 

Researchers used imagery taken from satellites developed by Maxar, a US space tech firm, including WorldView‐3, which currently orbits 372 miles (600km) above the Earth.   

As it sweeps across the land, the satellites can collect more than 5,000 km2 of imagery every few minutes, eliminating the risk of double counting. 

Where necessary – for instance, when land is obscured by clouds – the process can be repeated the following day, on the satellite’s next revolution of the Earth. 

‘This unobtrusive technique requires no ground presence so the animals are not disturbed, and human safety is not put at risk during data collection,’ said Maxar in a statement. 

‘Previously inaccessible areas become accessible, and cross-border areas – often crucial to conservation planning – can be surveyed without the time-consuming requirement of obtaining terrestrial permits.’

Using the highest resolution satellite imagery currently available – WorldView 3 – from Maxar Technologies, researchers have detected elephants from space with comparable accuracy to human detection capabilities

Satellite monitoring eliminates the risk of disturbing animals during data collection and ensures humans are not hurt in the counting process. 

It also makes it simpler to count animals moving from country to country, as satellites can orbit the planet without having to worry about border controls or conflict zones.

Currently, the most common survey technique for elephant populations in savannah environments is aerial counts from manned aircraft. 

However, observers on aerial surveys can get exhausted, be hindered by poor visibility and otherwise succumb to bias. Aerial surveys can also be costly and logistically challenging.    

Although this study is not the first to use satellite imagery and algorithms to monitor species, it’s the first to reliably monitor animals moving through a heterogeneous landscape – in other words, a backdrop that includes areas of open grassland, woodland and partial coverage.

‘This type of work has been done before with whales, but of course the ocean is all blue, so counting is a lot less challenging,’ said Dr Isupova. 

‘As you can imagine, a heterogeneous landscape makes it much hard to identify animals.’

African elephants were chosen because they are the largest land animal and therefore the easiest to spot. 

However, Dr Isupova is hopeful that it will soon be possible to detect far smaller species from space.

‘Satellite imagery resolution increases every couple of years, and with every increase we will be able to see smaller things in greater detail,’ she said.

‘Other researchers have managed to detect black albatross nests against snow – no doubt the contrast of black and white made it easier, but that doesn’t change the fact that an albatross nest is one-eleventh the size of an elephant.’

This kind of tech has great potential to support conservationists in their plight to protect biodiversity and to slow the progress of the sixth mass extinction – the ongoing extinction event triggered by human activity.

‘We need to find new state-of-the-art systems to help researchers gather the data they need to save species under threat,’ said Dr Isupova. 

The method and results have been detailed further in the Journal of Zoology.

Lire la suite:

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.

Articles récents

Suivez-nous sur les réseaux sociaux