Researchers at the University of California, Santa Barbara (UCSB) are using machine learning to identify and classify wildlife caught on camera to help ecologists.
Their "Where's the Bear" initiative uses a multitier, cloud, edge, and sensing system that combines innovations in machine learning-based image processing to automatically classify animals in images captured by remote, motion-triggered camera traps.
To build a training dataset, UCSB professors Chandra Krintz and Rich Wolski overlaid thousands of stock photos of wildlife onto background images from camera-monitored watering holes at UCSB's Sedgwick Ranch Reserve. The team used these synthetic images to train their device to automatically and accurately identify and classify animals, helping scientists aggregate and analyze more than 1 million images dating back years.
"Hopefully...we can eventually go to all the reserves in the UC system and help them, their researchers, and their students answer interesting questions and see how far we can push this," Krintz says.
From The UCSB Current
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found