Communications of the ACM
New Shark Spotter Algorithm Lets Surfers Take to the Break With Confidence
Researchers at the University of Technology in Sydney in Australia have developed the new Shark Spotter algorithm, which uses video footage streamed from drones to detect sharks and alert swimmers.
Credit: Guillaume Funfrock
Researchers at the University of Technology in Sydney (UTS) in Australia have developed Shark Spotter, an algorithm that uses video footage streamed from drones to detect sharks and alert swimmers.
The researchers say the algorithm is 90% accurate in distinguishing sharks from dolphins, rays, whales, and other marine life.
They note the algorithm's accuracy is a significant improvement over human spotters using their naked eye from helicopters in flight, which are 18% accurate, or those spotting from fixed-wing aircraft, which are 12% accurate.
They developed the app by flying drones over coastal waters and capturing about 8,000 images. The team then created the algorithm using computer systems modeled after the human brain and nervous system.
"The system efficiently distinguishes and identifies sharks from other targets by processing video feeds that are dynamic as well as images where objects are static," says UTS professor Michael Blumenstein.
From Digital Trends
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found