[argyllcms] Re: bin/average: averaging and possible outlier elimination for three or more .ti3 sets?

  • From: Craig Ringer <craig@xxxxxxxxxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sat, 29 Aug 2009 19:41:43 +0800

On Sat, 2009-08-29 at 12:25 +0100, Alastair M. Robinson wrote:

> > Also, is there any good built-in way to eliminate outliers in the .ti3
> > files, or will I need to roll my own ? For that matter, is it wise to do
> > outlier elimination at all?
> While I've not used an i1 myself, I've heard it said that it's not just 
> wise but vital.  Having said that, you don't want to eliminate process 
> variation - just misread patches.

OK, so that's definitely something I need to look into if (as I plan to)
I'll be trying to improve my results into something I can use in

> > I'd try to add basic outlier elimination for when it has three or more
> > inputs, with the outlier elimination threshold shrinking as the number
> > of input files grows. At this point I'm thinking that any sample more
> > than three (maybe even two) standard deviations from the mean is
> > probably a reasonable candidate for outlier elimination.
> So where you have an error spike you'd prefer to cull that sample from 
> all files rather than pick one to use?

Nope. I'd like to replace it with the mean of the valid readings of that
input, (or, really cull that particular input so it has no effect on the
mean *or* dE) . If I replaced it with another selected input I'd be
biasing the results in favour of the input value I happened to select.

Craig Ringer

Other related posts: