🚀 go-pugleaf

In article <Bz2n91.J3M@ie.utoronto.ca>
  green@ie.utoronto.ca (Marc Green) writes:
>It is the "objective"
>measures which are arbitrary and ad hoc. Many studies have shown that 
>mean squared error and similar measures are very poor predictors of
>perceived quality. And since images are created for humans to view, the
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>only empirically valid measure of quality is what human observers say it
>is. 

Careful....  This is a pretty limited idea of what images are good for!
Many branches of imaging science, mine among them, do quantitative analysis
of the images coming out of all the fancy algorithms.  Granted, good
quality metrics are hard to come by, and by necessity we aren't all that
sophisticated about them, but a decent error analysis on even a model image
is *extremely* valuable.  What it "looks" like to a human observer can be a
lot less important than maximum error.

-- 
| Daniel Briggs  (dbriggs@nrao.edu)                      |  USPA B-14993
| New Mexico Tech / National Radio Astronomy Observatory |      DoD #387
| P.O. Box O / Socorro, NM 87801   (505) 835-7391        | 
Support the League for Programming Freedom   (info from lpf@uunet.uu.net)