One-shot gigapixel camera offers a future beyond flat sensors
A gigapixel camera developed for the US Department of Defense's research agency (DARPA) provides an insight into the challenges that will need to be overcome to offer super-high-resultion cameras. A team from Duke University, Durham, North Carolina, has described its 960 megapixel (0.96 gigapixel) 'AWARE-2' camera in a letter to scientific journal Nature. The team says small, efficient electronics are the key to being able to miniaturize the camera, which currently sits in a 0.75 x 0.75 x 0.5m frame.
The team developing the AWARE-2 have considered scalability be one of the key considerations while designing it. Conventional camera designs will remain limited to megapixel resolutions because the small apertures they use are limited by diffraction softening. Scaling the design up and using larger apertures ends up limiting the system because lens aberrations increase, so you remain restricted to the megapixel scale, says team leader Professor David Brady.
This led to a design that arranges a series of cameras in a hemispherical arrangement, pointing at a single, spherical lens that the team have dubbed the 'gigagon.' The use of a single lens avoids the cost and complexity of having specialist optics on each sub-camera, while the curved design scales more easily than a flat array of cameras. The team believes this approach would continue to work for up to 50GP cameras.
|A diagram showing the hemispherical arrangement of the sub-cameras, and their relationship to the 'Gigagon' main lens (top right).|
The current design is made up of 98 individual 14MP sub-cameras, with focus and exposure set individually for each sub-camera. All the sub-cameras are exposed at the same time, meaning you don't have the problems of movement that occur in conventional, scanning gigapixel images. HDR techniques are then used to combine all the different 8-bit exposures into a single 32-bit file, which is then tone-mapped back to an 8-bit image that can be displayed.
|An image highlighting the contribution made by each sub-camera.|
The AWARE-2 weighs 93kg and captures a 120x42 degree field of view but the current design allows for a maximum of 220 sub-cameras to be installed, (giving a 120deg circular field of view). With all the cameras installed the overlap between sub-camera's images and elimination of poorly illuminated data would cut its theoretical 3 gigapixel capability back to around 2 gigapixels - around the number that its 16mm aperture would diffraction-limit it to, the group says.
The team says the image quality of the current camera is reduced by the use of injection-molded plastic relay optics in front of each sub-camera but that they believe this restriction can be overcome with the use of high-refractive-index plastics to improve this performance.
However, it's the electronics that the group says currently prevents the camera being made smaller - the optical system accounts for just 3% of the camera's volume - with the rest of the space taken up by the associated electronics and cooling required to dissipate the 430W expended every time the camera takes an image. With smaller, more efficient electronics, hand-held gigapixel cameras may become an everyday reality, they say.
|An example image, shot in Seattle, showing various crops taken from a 0.96GP image|
|Valley by the light of a blue moon by cjf2|
from Down in the Valley
|Lake Erie Stone Pier by yobbyt|
from Dock or Pier