AI method captures ecotourism photos to monitor remote animal species

Editors' notes

This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

peer-reviewed publication

trusted source

proofread

A team of computer scientists, ecologists and statisticians at Stony Brook University, working with a colleague from the U.S. Geological Survey, has found that it is possible to use AI applications to find images captured by ecotourists for use in studying remote animal species.

In their study, published in the open-access journal PLOS ONE, the group used a Segment Anything Model (SAM) to analyze imagery of Adélie penguins captured by ecotourists visiting the Antarctic.

In recent years, Adélie penguins have come to be known as a sentinel species, as they respond to changes in their environment due to global warming. But they are also a remote species, which means they are rarely studied in their environment.

The research team noted that two changes have occurred in recent years that could prove useful in studying the penguins. The first is the accessibility of cameras on smartphones. The second is the boom in ecotourism, comprising tours of remote places for average people. Ecotourists take a lot of pictures and/or videos of Adélie penguins.

The team concluded that a method to find such imagery and place it in context would be valuable for scientists, who could track the movements of colonies of the penguins over time. This would help them to learn more about the penguins, and also about climate changes occurring in the Antarctic.

The researchers developed and used a digital elevation model to align and register ground-based and satellite imagery. They then used the AI- based SAM for identification and segmentation purposes—it allowed for using context in the images to place the location of the penguins in their habitat.

By placing imagery in context, the researchers were able to track movements of colonies over time. They found that the penguins move around based on weather conditions, which gives clues about environmental warming. The team was also able to monitor colony size, which is affected by health conditions and the geographical conditions in which the penguins were living.

More information: Haoyu Wu et al, Penguin colony georegistration using camera pose estimation and phototourism, PLOS ONE (2024). DOI: 10.1371/journal.pone.0311038

Journal information: PLoS ONE

© 2024 Science X Network