Optimal column density measurements from multiband near-infrared observations

We consider from a general point of view the problem of determining the extinction in dense molecular clouds. We use a rigorous statistical approach to characterize the properties of the most widely used optical and infrared techniques, namely the star count and the color excess methods. We propose a new maximum-likelihood method that takes advantage of both star counts and star colors to provide an optimal estimate of the extinction. Detailed numerical simulations show that our method performs optimally under a wide range of conditions and, in particular, is significantly superior to the standard techniques for clouds with high column-densities and affected by contamination by foreground stars.