Preparing Red-Green-Blue Images from CCD Data

We present a new, and we believe arguably correct, algorithm for producing red-green-blue (RGB) composites from three-band astronomical images. Our method ensures that an object with a specified astronomical color (e.g., g-r and r-i) has a unique color in the RGB image, as opposed to the burnt-out white stars to which we are accustomed. A natural consequence of this is that we can use the same colors to code color-magnitude diagrams, providing a natural "index" to our images. We also introduce the use of an arcsinh stretch that allows us to show faint objects while simultaneously preserving the structure of brighter objects in the field, such as the spiral arms of large galaxies. We believe that in addition to their aesthetic value, our images convey far more information than do the traditional ones, and we provide examples from Sloan Digital Sky Survey (SDSS) imaging, the Hubble Deep Field (HDF), and Chandra to support our claims.