• Main Menu
Main Menu
Menu

Drone elephant recognition app

Dec. 18, 2019

MSU computer science students create agile automation app to save African wildlife

An innovative app by MSU students is working to curb the killing of elephants in Africa.
An innovative app by MSU students is working to curb the killing of elephants.

Africa loses around four elephants an hour and five rhinos a day.

To defend against poaching and protect the lives of wild elephants and rhinos, South African conservationists fly drones over vast land masses in protected areas to monitor herd movements.

Drone pilots use predictive tracking to maximize the chances of flying over elephants, which then allows rangers to be deployed against poachers.

The challenge: drones capture only a fraction of video footage that is usable intelligence on the creatures’ whereabouts. Then, thousands of hours of footage has to be manually reviewed, cataloged, and made actionable to keep the animals safe.

Evolutio, a group of technology professionals, asked Michigan State University computer science seniors to develop a drone elephant recognition and tracking application for the nonprofit organization Elephants, Rhinos & People (ERP). The goal -- monitor elephants on the Dinokeng Reserve in South Africa using agile automation and machine learning capabilities.

​  Winners for the Design Day Urban Science Sigma Award for best overall capstone experience in computer science are (front row, left) Mike DeRise of Urban Science; Rei Doko, Tyler Lawson, Kunyu Chen, and Jordan Cobe of Evolutio; (back row) Jeremy Arsenault, Nic Wiggins, Bob Dyksen of Evolutio; and Bill Bye of Urban Science.
Team Evolutio includes (front left) Mike DeRise of Urban Science; Rei Doko, Tyler Lawson, Kunyu Chen, and Jordan Cobe of Evolutio; (back) Jeremy Arsenault, Nic Wiggins, Bob Dyksen of Evolutio; and Bill Bye of Urban Science.

So a five-student team from the Department of Computer Science used their senior capstone project in the College of Engineering to create a state-of-the-art application to help. The app uses an automated object detection system that removes the need to manually analyze video footage, saving ERP hundreds of hours of time. ERP began using the app to analyze actual drone footage starting in fall semester 2019.

Team Evolutio presented its drone elephant recognition application as its senior capstone project during Design Day on Friday, Dec. 6, 2019 – and won the Design Day Award for Best Overall Capstone Experience Project in computer science. The Urban Science Sigma Award was presented to students Rei Doko, Tyler Lawson, Kunyu Chen, Jeremy Arsenault and Nic Wiggins.

Evolutio is now gearing up for a second phase of the ERP Elephant Drone Recognition project with a new team of Spartan Engineering students. Phase two in spring semester 2019 will have improved, targeted flight path data of elephant movements, which will allow even better training of the machine learning models in the application.

This rescue effort is just beginning. See details on this joint international effort in a video and blog.

Story courtesy of Evolutio.