The purpose of the MISR experiment is to acquire information of Earth at many different viewing angles. As we have nine cameras to do this job, we can take continuous images of Earth at 70 degrees, straight down (we call this a nadir view), and at a range of angles in between.
These pictures are not so much used to "look" at Earth as we would a movie. Rather, the measure of light which falls on any given camera is used to learn something about Earth underneath. For example, we can learn how much dust is in the atmosphere, how light is scattered from both the sides and tops of clouds, and how the land surfaces direct light in characteristic ways depending on the type of vegetation, soil, or nature of the urban development. Thus, although we do take pictures of Earth, our main purpose is to deliver information on Earth properties to the science community. They then track how these properties change globally, and through seasons and the years to come.
As we have said, in order to have scientific value, we must be able to record the amount of light falling into each of the nine MISR cameras - it is not enough to just create a pretty picture. We know the amount of light hitting a camera at any instant of time from a process we call "instrument calibration". Once the instrument is calibrated, we have knowledge on how to change the instrument output signal, related to its current output, to numeric values representing the light that hit the camera.
The MISR cameras have been calibrated, prior to launch, at the Jet Propulsion Laboratory assembly room. This was done using a giant 65" light box, containing 25 lamps. Here the output of the light box was known, and the MISR camera output electronic signal measured.
Many other preflight tests were used to make sure the MISR instrument was fully functional. The image below shows a target used to verify MISR image acquisition. The target was 4 x 6 feet in size. The target was moved across the image on a crane, to simulate spacecraft motion. When through, we had our first pictures acquired by MISR.
On-orbit, the MISR cameras are calibrated every month. To do this, a clean white plate is moved in front of the cameras. As we know how much sunlight comes from the plate, we can calibrate the cameras in space.
Another technique is to make use of MISR images over desert areas. The light from these sites can be measured by our ground crew. Averaging results from these various calibration techniques gives us both the MISR calibration constants, as well as our uncertainty in these numbers.