Photo courtesy of Society for Science
Despite conservationists’ efforts, animal poaching continues to devastate vulnerable species. So, when New Yorker Anika Puri came across ivory jewelry at a market in India four years ago, she felt inspired to do her part in stopping elephant hunting. The solution: she invented a low-cost machine learning software that can detect poachers in real time with 91% accuracy.
Discovering the numerous ivory objects in Mumbai was the catalyst for her project. “I was quite taken aback because I always thought, ‘Well, poaching is illegal, how come it really is still such a big issue?’” she says about the incident. So, the 17-year-old delved into the poaching numbers and discovered that Africa’s forest elephant population declined by about 61% between 2002 and 2011, with numbers that continue to drop.
Poachers are usually detected by drones; however, Puri noticed the success rate could be significantly higher. “I realized that we could use this disparity between these two movement patterns in order to actually increase the detection accuracy of potential poachers,” she explains. As a result, Puri spent two years developing her solution: a machine learning software named ElSa (an abbreviation for Elephant Savior). It analyzes the movement patterns of humans and elephants in thermal infrared videos and is four times more accurate than the existing detection methods. Even better, the software can be used with low-cost cameras, eliminating the need for high-resolution thermal cameras.
Puri presented her project at the Regeneron Internation Science and Engineering Fair, winning the $10,000 Peggy Scripps Award for Science Communication and first place in the earth and environmental sciences category. “It’s quite remarkable that a high school student has been able to do something like this,” Jasper Eikelboom, an ecologist at Wageningen University in the Netherlands, comments. “Not only the research and the analysis but also…being able to implement it in the prototypes.” Puri will be attending MIT in fall 2022 with hopes to expand her project to protect other endangered animal species.
A 17-year-old from New York named Anika Puri invented a low-cost machine learning software that can detect elephant poaching with 91% accuracy.
Photo courtesy of Society for Science
It analyzes the movement patterns of humans and elephants in thermal infrared videos and is four times more accurate than the existing detection methods.
h/t: [Smithsonian]
All images via Society for Science.
Related Articles:
Meet the All-Female Anti-Poaching Team Changing the Face of Conservation in Africa [Interview]
Gorillas Pose for a Selfie with Virunga National Park’s Anti-Poaching Rangers
Watch Olive the White Rhino Give Birth to a Healthy Baby Calf