06 March 2014
There are still many things that humans can do better than robots, such as seeing the world around us. Robot eyes are usually based on the kind of technology used in digital cameras, capturing an image in a single picture. But human eyes don’t work like that. Instead, we gather information from many small, overlapping pictures, which the brain then builds into a single image. Computer researchers are now trying to copy this approach in order to make robots with more precise vision. The panels on the right represent what a robot with camera-like eyes would ‘see’ when looking at either a regular (top) or random (bottom) pattern, with the blue and pink colouring highlighting areas lacking in detail. But the images on the left are ‘seen’ using overlapping pictures, revealing much more fine detail. This technique could change the way the robots of the future sense the world.
Written by Kat Arney
Image courtesy of Kevin Earland and colleagues
Aberystwith University, Wales
Originally published under a Creative Commons Licence (BY 4.0)
Research published in PLOS One, January 2014