I imagine the answer is "maybe, but probably not".
The big problem I see is that it seems that when most people say "halftone", they don't mean halftone at all.
they mean dithering.
So without any information, I'm left unable to know which problem is being described.
If you have a halftone image, then is it digitally halftoned, or is it a photo or scan of a print?
If you have a dithered color image then you might be able to tell some things, but I doubt an exact answer is obtainable. For a black/white reduction of a grayscale image, you might be able to get away with more, but for color images, you're left trying to derive one or more coefficient matrices and determine the underlying quantization method and the color space in which quantization was performed -- all blindly without a reference. I don't think there's enough information to do that.
With that in mind, maybe it also matters what specific information is required. Is it actually necessary to find the size of the coefficient matrix and all of its values, or is it sufficient to just say "this image uses ordered dithering"? Also, is it necessary to do this programmatically for multiple images, or is it sufficient to show one image to someone and get an educated guess?
