6/28/2012 4:00 AM
Get ready for another major leap forward in digital photography—one that could transform television news in the future. A military-funded project at Duke University has developed an experimental video and still camera that allows the photographer—after the original photo is taken—to zoom in on parts of the image in extraordinary detail.
News of the camera, first reported by the Wall Street Journal and in the scientific journal, Nature, is a major development in digital camera technology. It collects more than 30 times as much picture data as today’s best digital devices, and can produce a video or still image with a billion pixels. That’s five times as much detail as can be seen by a person with 20/20 vision, the Journal reported.
Called Aware-2, the Duke-developed camera is still a prototype. The current version is the size of two microwave ovens and weighs about 100 pounds. It generates a lot of heat and needs cooling. It also takes about 18 seconds to shoot a single frame and record the data on a disk.
A key breakthrough to the Duke camera is its spherical lens, a design that dates back to the late 19th century. Although effective spherical lenses already exist—in the human eye, for one place—researchers have long found it tricky to accurately focus images using lab-made versions, the Journal reported.
The Duke team overcame the challenge by installing nearly 100 micro-sized cameras, each with a 14-megapixel sensor, on the outside of a small sphere about the size of a football. The configuration creates nearly 100 separate images, each accurately focused. A computer connected to the sphere then stitches the images together to create a composite whole.
The current prototype takes only black-and-white pictures. The Duke group will finish building a 10-gigapixel color version by year-end and then will construct a 50-gigapixel device. They hope to begin manufacturing industrial-type gigapixel cameras on a limited basis next year. But the scientists estimate it will take several years before a hand-held version of the technology becomes available.
Assuming the camera can be developed into a usable size for new photographers and others, users could simply shoot any scene and then later re-compose the image and zoom into any part of the detail. Due to the large amount of data captured, any part of the frame would offer superior detail for still images.
David Brady, an optical engineer at Duke, told the Journal that when his team used the device to photograph the Seattle skyline, they were able to zoom in and read the “in” and “out” signs written on a parking garage a half-mile away. Similarly, he said, if the camera were used to take video images of a tennis match, say, the viewer could zoom in on a player, or on someone at the far end of the stadium, and see both images with equal clarity.
Although high resolution cameras exist for telescopes and other devices, the Duke cameras represents the “first cut” at making gigapixel cameras for general use, Shree Nayar, a computer-vision researcher at Columbia University in New York, told the Journal. The challenge, he said, is to shrink the electronics and reduce the amount of power the system requires.
Another issue is the enormous amount of data such a camera would generate, the report said. Camera makers would have to design software to process all the data since it would be beyond the capacity of most of today’s computers.
The Defense Advanced Research Projects Agency, part of the U.S. Department of Defense, invested about $25 million in the research. It wants a high-resolution camera for aerial and land-based intelligence, the Journal reported.
See a video of the technology here.