Over a period of several months in 2005 and 2006, a half-dozen artists converted a shuttered F-18 aircraft hanger into the largest camera ever made. To darken the hanger the artists used 24,000 square feet of viscuine, 1,300 gallons of foam gap filler, one and a half miles of gorilla tape, and 40 cans of black spray paint. They then suspended a 107-foot by 31-foot piece of gelatin-silver coated muslim from the hanger’s ceiling and used a quarter-inch diameter pinhole to expose the giant ‘film’ with light from the nearby surroundings. The camera did not use a lens to form its image.
This record-setting image is indeed impressive, but the basic idea of capturing images through a pinhole like these artists did with their camera is relatively simple.
If the rays from a small beacon of light fall upon a flat device like a chemically-coated film or optical detector, each position on the film or detector will sense and record an equal amount of light. Furthermore, if light originates from two different beacons the detector will still be lit uniformly across its surface.
The primary job for a camera, then, is to isolate the rays that originate from unique positions in a way that forms a faithful image on the detector. Two light sources in a scene should give rise to two bright spots on a detector, and three light sources in a scene should give rise to three bright spots on a detector. One way to accomplish this is by filtering the light rays with a simple pinhole in a screen.
To make a pinhole camera we simply place an opaque screen somewhere between the detector and the scene. Because the screen will filter some of the light rays from the sources, the detector’s surface will not be uniformly lit by the light. Instead, the detector will sense light from a two-source beacon in two localized regions.
There are two counterbalancing drawbacks in the use of a pinhole camera. Although a pinhole camera does an excellent job of isolating the light rays that originate from different locations in the scene, much of the light from the scene is blocked by the screen, and very little light finds its way to the detector. The exposure time to collect an adequate amount of light with a pinhole camera, then, can be several minutes or hours. We can capture more light by increasing the size of the pinhole, but, in doing so light from one location in the scene will spread over a larger region in the detector creating what is known as an image circle of confusion.
By widening the pinhole to capture more light we also widen the circles of confusion and make the image more blurry. If, for instance, we make the pinhole too wide, the image of two light sources might blur into one large image on the detector.
A natural alternative to the pinhole idea is to place a device in the hole – or aperture as it is called – that will bend each ray that originates from a particular position in the scene to the same location on the detector. The delightful device that accomplishes this task is a lens, and Snell’s law provides the key to understanding how it works.
© 2011 Timothy Schulz