The AI system helps cars detect hidden objects in corners of roads

A team of researchers from Princeton University, Stanford University, Rice University and Southern Methodist University recently designed a powerful laser system using a laser pointer,media reported. The beam bounces through a visible wall onto a hidden object behind the corner, and then the beam bounces back from the object and falls against the wall.

The AI system helps cars detect hidden objects in corners of roads

At this point, the beam creates an interference pattern on the wall, called a pattern. Images of hidden objects are encoded in this pattern, and the researchers use a deep learning algorithm to expand complex calculations for real-time reconstruction.

Currently, researchers have been able to successfully read numbers and letters that cannot be seen directly through sight.

“Our non-line imaging systems provide unique high resolution and imaging speeds compared to other methods,” said Christopher A. Metzler of Stanford University and Rice University, who led the study. He added that these properties make it possible for applications that would otherwise be impossible, such as reading the license plates of hidden vehicles while driving or reading the badges worn by people walking on street corners. “

The study is believed to have been published in Optica.

“Non-line imaging has important applications in medical imaging, navigation, robotics, and defense,” said co-author Felix Heide of Princeton University.