[Radiance-general] HDR Image Accuracy Problem

Fitzpatrick, Halley D HDF2 at pge.com
Thu Aug 25 03:13:35 CEST 2005


Hello! This is my first post/question...

I have been using Radiance on and off for about a year and a half, and
now I am delving into high dynamic range images. I am trying to create
accurate and precise false color luminance maps from HDR images. I took
a series of pictures each with 5 exposures, and found out quickly that
was not enough for what I am trying to accomplish. The luminance values
(as displayed by photosphere and the false color I created) looked about
right, with the exception of particularly bright portions of the image,
which were damped by not having enough exposures. The extreme values are
quite important to me.

I decided to start over, and took 11 exposures of a scene: 5
underexposed, 5 overexposed, 1 properly exposed (each a full f-stop
apart) with a Nikon D100. The proper exposure was determined by the
camera using matrix metering. The most underexposed image had a small
portion that seemed to be washed out with white, but I can't take any
more underexposed due to my camera (I don't think this is causing my
problem, anyway). I created an HDR image using photosphere, and chose
'save new response' and 'align images', leaving the rest of the fields
unchecked. Clicked go, and created the hdr image. The image _looks_
good, but I think that the values are likely quite a bit off.

I measured several values in the scene with a luminance meter (recently
calibrated Minolta LS-110) and all of them are 15-20 times less than the
values suggested by photosphere. The values I was seeing with 5
exposures were closer to the actual luminance meter values, but I have a
feeling now, that was a coincidence. 

I will be happy to provide the jpegs and/or the hdr that I created if
they will serve to discover my problem.

Any help or insight would be greatly appreciated. Thank you!

Halley Fitzpatrick              [ Program Engineer ]
The Pacific Energy Center       [ p. 415 972 5238  ]
851 Howard Street               [ f. 415 896 1290  ]
San Francisco, CA 94103         [ e. hdf2 at pge.com  ]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://radiance-online.org/pipermail/radiance-general/attachments/20050824/813e9025/attachment.htm


More information about the Radiance-general mailing list