Teenager Invents AI Software Against Poachers

A 17-year-old American girl from New York came up with a system to accurately identify and catch elephant poachers using movement patterns, thus preserving the growingly endangered giant species.

Anika Puri was inspired to do something about the alarming declining rate of Africa’s forest elephant population – by 62% between 2002 and 2011. Upon noticing that the images captured by drones lacked accuracy, “I realized that we could use this disparity between these two movement patterns in order to actually increase the detection accuracy of potential poachers.”

And from this realization, she created ElSa – which stands for Elephant Savior – a low-cost prototype of a machine-learning-driven software capable of analyzing movement patterns in thermal infrared videos of humans and elephants. Four times more accurate than existing detecting methods, ElSa eliminates the need for pricy high-resolution thermal cameras. Instead, it uses the $250 FLIR ONE Pro thermal camera with 206×156 pixels resolution which can be plugged into an iPhone 6. Both camera and iPhone are fixed to a drone and real-time inferences are produced. Data from the Benchmarking IR Dataset for Surveillance with Aerial Intelligence became the basis of the model. Puri hopes for her technology to be implanted in national parks in Africa, and she intends to extend her movement pattern research to other endangered animals such as the rhino. “Research isn’t a straight line. That has made me more resourceful. It also helped me develop into a more innovative thinker. You learn along the way.”