I have a question about the procedures described in the first half of the tutorial, "A New Approach to Combination of Broadband and Narrowband Data." I understand the first-pass iteration, but am not sure about the second and subsequent. I form the first continuum map, smooth it, then multiply it by the narrowband (Ha) to get the refined broadband (Red) image. Now the question: is this "refined broadband" image the one that gets smoothed in the next iteration? I suspect so, because if we go back to the first equation in the article and divide by the narrowband to get a continuum map, then we have just multiplied and divided by the same thing...that cannot be correct. So I guess I should now apply another smoothing to the "refined broadband" and multiply that result (again) by the narrowband, to get something like a refined(refined(broadband)), which I use as if it were a "normal" red channel in RGB combine?
If my refined(refined(broadband)) procedure is correct, I expect that the appropriate second (and subsequent) smoothings will likely be much milder than the first...right?
Your help greatly appreciated,
Alex
