(3) the 'Vibrance Mask's white represents 0% saturation, black represents 100% saturation plus the 'Subtract-ed' low saturation ColorRange selection, and the grays map the original saturation gradient from low to high saturations.
Should read:
(3) the 'Vibrance Mask's white represents 100% saturation, black represents 0% saturation plus the 'Subtract-ed' low saturation ColorRange selection, and the grays map the original saturation gradient from low to high saturations
A final report from me on the matter. Again, my thanks to the generous replies to my original posts. Based on the feedback from the posts I compared a method in PWP with the vibrancy slider in the Lightroom 2.3. In PWP I extracted the HSV-S channel and converted to 8-bit to make mask. I then applied Den's RGB color curve (http://www.ncplus.net/~birchbay/3tone/3tone.htm#Add2) in HSV-S color space with the mask applied to the image. Black slider at 0 and white slider adjusted for taste. The result is much more pleasing to me than LR which tends to saturate lighter colors (whites and greys) with a yellowish hue. Finally, this process has helped me understand color curves better, a transformation which has intimidated me in the past.
Thanks for the procedure for building a 'vibrance' mask.
However, there is one thing that puzzles me.
As I undestand things the idea of the vibrance mask is to protect areas of high saturation AND skin tones during any transformation. If this is the case, shouldn't step (5) in "The Vibrance Mask:" procedure be Add rather than Subtact?
For the illustration below:
1 = the 'saturation mask' [extracted HSV-S]
2 = the 'skin color mask' [Mask Tool - Color Range]
3 = the 'vibrance mask' ['saturation mask' minus 'skin color mask']
image = Saturation transform - preserve neither, vibrance mask white = 50, vibrance mask black = 0