x
Breaking News
More () »

USF researchers using algorithm to detect 'hidden objects'

The researchers believe this technology could one day be used by self-driving cars to prevent crashes or by law enforcement to help detect threats.

TAMPA, Fla. — It is sci-fi meets reality at USF in Tampa, where researchers are developing an algorithm that allows them to see around hidden corners and through obstacles.

"We're trying to see around corners, and so we want to reconstruct a picture of something that we can't see directly." said Ph.D. doctoral candidate Robinson Czajkowski.

The algorithm is pages long and allows the researchers to take an image of a wall and use it like a mirror. 

In roughly 30 minutes to a few hours, the algorithm analyzes the shadow and color projected onto the wall and uses them to interpret the scene that would be before it. 

"Humans, we need to see shapes in order to put things together and understand what's, you know, what's there. But the computer, all it sees is numbers. So it's able to do the reconstruction just as well," Czajkowski said. 

Over the next decade, Czajkowski and USF Assistant Professor of Computer Science John Murray Bruce aim to increase the algorithm's speed and overall image quality. Eventually, they hope the technology is robust enough to be adopted by law enforcement and car manufacturers. 

"So law enforcement could be more prepared in those types of situations in understanding, for instance, the pose of the person and if they're welding something," Professor Bruce said. "In a hostage situation, you may be able to better understand how many people are in there." 

The work ahead of them is extensive and uncertain, but Murray Bruce said he's undaunted. 

"We're excited about our work and we do think we have something here. That's what drives our research. We will look at challenging problems and hope to solve them and make a huge impact in reality," said Murray Bruce. 

  

Before You Leave, Check This Out