A camera that can see five times better than a person with 20/20 vision is here. The resolution is 50 gigapixels, or 50,000 megapixels. Cameras used by professionals reach the 40-megapixel mark, and a typical point-and-shoot might have eight or 10. The amount of information the gigapixel camera captures in a frame is nearly as much as the amount of data stored on a PC's hard drive — all for one picture.
The group that developed the camera was led by David Brady, professor of electrical engineering at Duke's Pratt School of Engineering, along with scientists from the University of Arizona, the University of California — San Diego and Distant Focus Corp. The Defense Advanced Research Projects Agency (DARPA) also supported the research. The military has an obvious interest in better image processing and in decades past, has spearheaded the development of sophisticated imaging tools. But a camera like this is also useful for fields as such as astronomy.
Unlike conventional digital cameras that have a single image sensor, the gigapixel camera has 98, all working in unison. A computer puts all the disparate images together to create a single picture. Blowing up the image reveals details that the naked eye wouldn't initially see.
Michael Gehm, assistant professor of electrical and computer engineering at the University of Arizona, who led the research there, noted in a press release that to boost the camera's resolution, the team had to go beyond simply making the optical elements more complex. This time, the researchers improved the electronics and built a system of parallel processors. The camera lens focuses the light to each of the image sensors and the computer takes care of building the final image.
Processing all the information is so complicated that 97 percent of the physical camera is comprised of the electronics and computer; only about three percent of the camera itself is deveoted to the optical elements. Details of the new camera were published in the journal Nature.
Photo: An image from the gigapixel camera, showing enlargements of certain sections. Credit: Duke University Imaging and Spectroscopy Program