“Intelligent” goggles highlight obstacles for the visually impaired
By David Szondy
May 30, 2012
Researchers at the Universidad Carlos III in Madrid (UC3M), Spain, have developed a pair of “intelligent” goggles that make getting around a bit easier for partly-sighted people, by providing them with Terminator-style vision. Using a pair of cameras mounted on a virtual reality headset connected to a tiny computer, the device scans the area ahead of the wearer and displays information about the scene as color-coded outlines that convey the distance and shapes of objects that are difficult to otherwise see or interpret.
There are many devices available to help the blind, and more are rolled out every year. They range from traditional canes and guide dogs to speaking clocks and watches and tactile mobile phones. Since the 1960s, a lot of work has gone into ultrasound devices to provide the blind with a sort of personal radar for avoiding obstacles. In recent years, this has led to wrist-mounted sonic navigators, canes, vests and even smartphones, designed to help the visually disabled get around and interact with other people.
However useful these ultrasonic and similar devices are, they rely on hearing and touch to communicate with their users. Many people who are legally blind still retain a degree of sight, however. Their vision may be profoundly blurry or very restricted in area, but it is enough to provide some information about the visible world. Yet outside of very high-powered glasses, digital magnifiers or devices with giant displays, there aren’t very many bits of technology out there to help those who still can see somewhat.
Led by Professor Ricardo Vergaz of the Electronics Technology Department, the UC3M team designed the “intelligent” goggles to take advantage of the wearer’s remaining sight. The system consists of a pair of stereoscopic digital cameras mounted on either side of a virtual reality headset, with two digital screens in front of the wearer’s eyes in place of lenses. The cameras scan the field of vision in front of the headset, convert it to digital code and then feed this to a separate computer package. The computer then runs an algorithm developed by the team, that determines the distance and outline of any objects seen. What the cameras scan is displayed on the headset’s screens and information about the objects is conveyed to the wearer by overlaying them with color-coded silhouettes.
“It detects objects and people who move within the visual field that a person with no visual pathologies would have," said Professor Vergaz. "Very often the patient does not detect them due to problems of contrast. The information regarding depth is what is most missed by patients who use this type of technical aid.”
Currently, testing is underway on a representative sample of patients who are suitable for using the device, and results are expected by the end of 2012. These results will be used to improve the performance of the “intelligent” goggles. Meanwhile, the team is concentrating on improving the ergonomics of the device by shrinking the goggles to something more practical and comfortable.
Source: Universidad Carlos III in Madrid