Climate change, fires, desertification, Parkinson’s, food waste, fishing, agriculture, bees, earthquakes, elephants, swallows, children with autism, oceans, … are some examples that we have dealt with in this Blog, and that thanks to artificial intelligence (AI) will be able to improve their prospects for life, prediction, or even salvation. Today we add to this list the snow leopards, one species, one more, in danger of extinction , but which can also be protected thanks to technology.
Protecting snow leopards from extinction begins with knowing how many there are today, something difficult considering that they live in cold and arid lands, that they can travel hundreds of kilometers and that they are so elusive that even experts nickname them ” the ghosts of the mountain ”. However, advances in artificial intelligence can play a key role in monitoring the health of these felines and their environment.
Snow Leopard Trust, a Seattle-based non-profit organization, aims to safeguard this endangered animal through better scientific understanding of its behaviors , needs, and threats. For this, this association has been using cameras equipped with heat and movement sensors, a solution that has been affected by incidents such as avalanches, floods or because of the animals themselves that move them.
Installed in the habitats of Central and South Asia, these cameras generate hundreds of thousands of photos, many of them of goats, camels, horses, and even swaying grass. In short, anything that activates the camera sensors. Thus, classifying images into photos with and without leopards was traditionally a long and tedious manual task.
But this problem has ended since the Snow Leopard Trust used Microsoft’s AI to analyze and classify images in a database, doing what used to take 10 days in 10 minutes. “Analyzing all these images used to occupy a considerable part of our time,” says Koustubh Sharma, a renowned regional ecologist with the non-profit organization that has been studying these wild cats for 12 years and is also the international coordinator of the Global program. Snow Leopard & Ecosystem Protection. “By automatically analyzing the images and creating a database for us, our small team is given the time to do more surveys and collect better data.”
Image classification system helps Sharma and his team save time, access critical data faster, and optimize resources for protecting snow leopards from poaching, the effects of mining, climate change and other threats.
The next step for the technology would be to automate the identification of these individual felines based on their unique markings. According to Sharma, currently no one really knows how many leopards there are, although the best available estimate is 3,000 to 7,000 . In addition, it also ensures that the existence of this species is important for humans because they are important ecological “thermometers” of the mountains where they live and that supply water to neighboring towns.
The presence or absence of snow leopards indicates if all is well in the ecosystem. By using cameras to study the snow leopard population, you are actually working to protect an entire ecosystem .
