Hi Yulia, 1. You need the "-o" option to pvalue, and the rest should work. The ".pic" file is the same as the ".hdr" file produced by Photosphere, a.k.a. "Radiance RGBE format" in the "Save As..." dialog.
2. The standard deviation for luminance is computed in the log domain, so what it gives you is a multiplier/divisor on the value rather than an absolute deviation. In other words, one standard deviation above is 1.1x the median, and one standard deviation below is median/1.1. 3. Averaging in a circular area is tricky and usually not worth the effort. There aren't any luminance meters that are accurate enough about their circles to worry about the difference between a circle and a square. If you must do it, it's a long and nasty-looking pcomb command. Best, -Greg > From: "Tyukhova, Yulia" <[email protected]> > Date: March 11, 2012 10:52:40 PM PDT > > Hello everybody, > > I have a couple of questions on HDR image analysis. > > 1. How could I extract the luminance values from image.hdr? > > With the pvalue tool I get RGB values. And in August 2007 thread Greg > mentioned this (to get L): > % pvalue -h -H -b render.pic \ > | rcalc -e '$1=$1;$2=$2;$3=179*$3' > But I would assume this is for .pic files. Is there a way how I can get it > from .hdr that was already calibrated in Photosphere? > > 2. How is a standard deviation calculated in Photosphere? What does 1.1x > mean? I would expect to see number closer to the luminance value displayed in > the image. > > 3. In June 2009 thread there was a discussion on averaging luminance values > withing the circular area. From anybody's experience, what is the easiest way > to get average L value from hdr image within the circle area? > > Thank you for all your help, > Yulia
_______________________________________________ HDRI mailing list [email protected] http://www.radiance-online.org/mailman/listinfo/hdri
