Blocking the lens, as the group has done, makes a tiny, fuzzy ring appear around an image. However, this ring is about a micrometer in diameter, and since most of the light sensors in cameras are only sensitive to a resolution of two micrometers, the ring is undetectable. “If you had a perfect detector with infinite resolution this would be a disaster,” says Ford. “But it works well for the kind of sensors that we’d find for [digital cameras].”
The UCSD research takes advantage of a well-known optical system, says José Saisián, professor of optical sciences and astronomy at the University of Arizona, in Tucson, but he admits that the design is unique. “I think it has some merit,” he says. “They took this idea, analyzed it well, and it can have some interesting applications.”
Ford acknowledges that there were some drawbacks to the group’s initial prototype. The first prototype had a limited depth of focus, for example, which meant that anything approximately two inches in front of or behind the focus point of the lens will appear blurry. However, he says, his team has explored different-shaped lenses that increase the depth of focus, and it has built successful camera prototypes. A second, smaller prototype, using a slice cut from a round lens, matched the depth of field of the conventional camera. Ford claims that the third generation imager, which is now being tested, will be even smaller.
When designing an embedded system choosing which tools to use often comes down to building a custom solution or buying off-the-shelf tools.