Using a mathematical algorithm and a video camera, University of Southern California (USC) researchers created pixels that are projected onto a headset to improve vision for people with retinal implants. The headset can assist with tasks such as navigation, route planning, and object finding.
Retinal implants are currently of low resolution, but they allow blind people to sense motion and large objects, thereby improving orientation for walking and enabling most users to read large letters. The algorithm will supply retinal implant users with more information when they are looking for a specific item, says USC's James Weiland.
The researchers mounted a video camera on a headset to collect real-world information in the view of the subject, then transformed the images into pixels that were displayed on a screen in front of the subject. The algorithms use intensity, saturation, and edge-information from the camera’s images to pick out the five most salient locations in the image. Additional directional cues are supplied via blinking dots at the side of the display.
To test the headset, researchers asked subjects to walk an obstacle course, find objects on a table, and look for specific items in a cluttered environment. The directional cues drastically lowered errors, head movements, and time needed to complete tasks.
From Institute of Physics
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found