Re: Does Image Magick have Tonemapping operators?
Posted: 2011-08-10T09:13:51-07:00
IM only has one way to do bilateral filtering using -selective-blur which does a gaussian blur with a simply gray level threshold to limit the range of graylevels. But I found that using it (in my limited tests) it left ringing artifacts. So I went back to a different -adaptive-blur, which did not leave such artifacts. I am sure this is not an optimal bilateral filtering, but at least it followed the other concepts as best I could understand from the Durand paper. It seemed to me to be better than my other methods. Best I could do by scripting.
Yes, I noticed the halo on the last image, but have no idea why it is there as this method was not dual process that separate the shadows and highlights as in my dualtonemap and duotonemap.
Again, it seems to me that if these psfto algorithms are open source c code, then perhaps the IM developers could fold at least one or two of the better ones into IM.
Quote from one of my old colleagues in image processing:
"90% of the result comes from the simplest 10% of the processing; it is the other 10% of the result that takes 90% of the effort"
Yes, I noticed the halo on the last image, but have no idea why it is there as this method was not dual process that separate the shadows and highlights as in my dualtonemap and duotonemap.
Again, it seems to me that if these psfto algorithms are open source c code, then perhaps the IM developers could fold at least one or two of the better ones into IM.
Quote from one of my old colleagues in image processing:
"90% of the result comes from the simplest 10% of the processing; it is the other 10% of the result that takes 90% of the effort"