Noise and Random Values
An important part of generating realistic images is to add appropriate levels of noise. When simulating objects, you may also want the actual objects being drawn to be random in some way. GalSim has a number of classes to help inject these kinds of randomness into your simulations.
Random Deviates describes how to generate pseudo-random numbers according to a variety of different probability distribution functions.
Noise Generators describes how to add noise to images according a a few different noise models.
Correlated Noise describes both how to add correlated noise to images and also ways to remove (or reduce) existing correlations from an image.