How Hubble/ACS color images are made from two black-and-white images:
ACS images are taken through color filters that isolate particular wavelengths in the visible/near-infrared region of the spectrum. These individual images are rendered in black-and-white but can be combined to make color images. The above three-part figure illustrates this process.
Panel a) is a single ACS V-band image of the object CXO-J141741.9, which appears to be a quartet of merging galactic fragments whose interaction has triggered a mini-quasar (bright point-source at upper left) and a starburst. The V filter captures yellow light near the middle of the spectral range that is visible to the human eye.
Panel b) is a similar view through the I filter, which is located redward of V towards the near-infrared. Panel c) is the color composite made from the two images. If a pixel is brighter in I (red) than in V (yellow), it is colored red in the color composite. If a pixel is brighter in V (yellow) than in I (red), it is colored bluish in the color composite. Pixels that are equally bright are colored gray (neutral). It is obvious that the mini-quasar is quite red, which is thought to be due to dust clouds surrounding it that absorb and redden the light. (Galactic dust grains absorb more blue light than red, and so objects appear reddened when partially screened by dust in much the same way that the setting Sun looks red when its light traverses a long path through the atmosphere.)
Color images like these are extremely valuable---not only do they map the presence of dust, as here, but in more typical galaxies they signal star-formation histories. Recently formed young stellar populations are blue, while old stellar populations are "red-and-dead."