Seeing through the clouds: Belvedere researcher uses AI to unlock aurora data

In 2008, Belvedere’s Jason Press stood in a darkened bathroom watching his father develop animal X-rays — a veterinarian’s makeshift darkroom, a boy fascinated by science.

“We had to go into the bathroom, turn the lights off, because that’s where the developing tank was,” he said. “It’s just like you develop an old-school roll of film. That stuff was so cool to me.”

Today, at 22, the Pepperdine University senior is still pulling images out of darkness — this time, the aurora borealis.

Press has spent the past year building an AI system that takes footage of a cloud-covered Arctic sky and recovers a clear view of the northern lights hidden behind it, rescuing scientific data that researchers had written off as lost. He presented the work at Pepperdine’s research symposium on April 10 — the latest milestone for an undergraduate project that has already produced a peer-reviewed publication.

“The path that I have is getting literally a camera picture with clouds, and making the clouds go away so that we can see the aurora behind it,” he said.

Read the complete story in our e-edition, or SUBSCRIBE NOW for home delivery and access to the digital replica.

Comment on this article on Nextdoor.