Researchers Create Method Using Fast Flickering LEDs to Impede Digital Image Capture

At many locations, prohibitions against digital photography are in place, either out of privacy concerns or to protect trade secrets. With the ubiquitous nature of smartphones with integrated digital cameras, enforcing such rules requires the cooperation of the potential photographer. However, some researchers at the University of California San Diego and the University of Wisconsin Madison have devised a method that hinders any efforts to take digital photographs or video. The researchers used LED lighting switching on and off very quickly to confound digital light sensors in digital cameras and video equipment. They dubbed the system LiShield.

Flickering LEDs at various frequencies results in digital images that have periodic breaks with distorted colors. While a digital camera or video device can see sense such fast flickering lights, people cannot. So, causing the LED lights to flicker at a particular frequency is an unobtrusive way to help enforce a no camera or video policy.

University of California San Diego and University of Wisconsin Madison -- Impact of on and off frequency on image distortion

University of California San Diego and University of Wisconsin Madison — Impact of on and off frequency on image distortion

On the other hand, according to the researchers, if a person is granted permission to photograph or take video of the area under the flickering lights, the digital light sensing technology can synchronize with the LED lights to create pieced-together images and video that overcome the distortions. To make it especially difficult for digital sensors to overcome, the researchers packed multiple frequencies into every image so synchronization requires knowing the pattern of on and off switching.  So, a proper image recovery method has to get data from multiple images to capture the entire on and off pattern.

Luxeon High Power
University of California San Diego and University of Wisconsin Madison-- multi-image recovery method for authorized and unauthorized users

University of California San Diego and University of Wisconsin Madison– multi-image recovery method for authorized and unauthorized users

The researchers applied the technique to a single LED light source, but they suggested that with the proper controls such periodic shifting of on and off lights could also be applied across an entire light system or in a particular area.

While not explicitly acknowledged as a possibility, in theory using such methods would not prevent the use of the LED lights for LiFi if the LiFi encoding and decoding takes into account the periodic off and on switching of the LED lights.

The researchers detailed the method in a paper which they presented at an ACM Mobicom 2017 conference.

Reference

S. Zhu, C. Zhang, X. Zhang, “Automating Visual Privacy Protection Using a Smart LED,”  formally presented at ACM Mobicom 2017 conference.