Photography

Turning image noise into a good thing

Turning image noise into a good thing
As the electric current of the strontium barium niobate crystal was altered, the noisy image regained clarity
As the electric current of the strontium barium niobate crystal was altered, the noisy image regained clarity
View 2 Images
Princeton engineering researchers Jason Fleischer and Dmitry Dylov with their imaging system
1/2
Princeton engineering researchers Jason Fleischer and Dmitry Dylov with their imaging system
As the electric current of the strontium barium niobate crystal was altered, the noisy image regained clarity
2/2
As the electric current of the strontium barium niobate crystal was altered, the noisy image regained clarity

Noise in images is generally held to be a bad thing, but engineers from Princeton University have used a nonlinear material to steal energy from image noise to reveal hidden or obscured objects. The engineers see the technology as potentially paving the way for improvements to radar systems, sonograms and stenography offering the possibility of allowing pilots to see through fog and doctors to look inside the human body without surgery

Jason Fleischer, assistant professor of electrical engineering at Princeton University, explained: "sometimes noise and signal can interact, and the energy from the noise can be used to amplify the signal. For weak signals, such as distant or dark images, actually adding noise can improve their quality."

In experiments, the engineers first passed a laser beam through a small piece of engraved glass to carry an image of lines and numbers to a receiver connected to a video monitor. Then a piece of translucent plastic was placed between the glass and the receiver to scatter the laser's light, making the received image so noisy as to be indecipherable. Finally, the engineers mounted a crystal of strontium barium niobate just in front of the receiver.

The nonlinear crystal mixed up the different parts of the received beam and allowed the signal and the noise to interact. Altering the electric current across the crystal allowed the engineers to fine tune the received image and clear up the scattered mess of light so that the lines and numbers reappeared.

"We used noise to feed signals. It's as if you took a picture of a person in the dark, and we made the person brighter and the background darker so you could see them. The contrast makes the person stand out," said co-author Dmitry Dylov.

Already known in various fields (from neuroscience to energy harvesting), stochastic resonance has not previously been used to bring clarity to images in this way. The engineers have developed a new theory for how noisy signals move through nonlinear materials which could provide the starting point for future developments that may allow pilots to see through fog and doctors to look inside the human body without surgery.

The research findings were published online last month in Nature Photonics.

No comments
0 comments
There are no comments. Be the first!