each_channel decorator

[Tony Yu]

On Sat, Oct 20, 2012 at 2:13 AM, Schönberger Johannes <
hannesschoenberger at gmail.com> wrote:

> Hm, I get the feeling that this decorator gets too complex and "magic"… Is
> this really a standard use case? If someone really wants to apply a filter
> to the luminance channel of an image, why not let them use ***2lab function?
>
>
> Johannes Schönberger
>

"Hidden" and "magic" may have been the wrong wording. Providing decorators
allows filter writers to provide a default RGB behavior. There doesn't
really need to be an `rgb_behavior` parameter at all (i.e. no user-settable
parameter). Regardless of what the filter writer decides, the user can
adapt the grayscale filter however he or she wants---either with utility
functions or explicitly with, e.g., rgb2lab.

My main hesitation with utility functions is that they make argument
passing a bit ugly. On the other hand, I think doing it explicitly is quite
involved (e.g. LAB: 1. convert to LAB, 2. grab the channel, 3. rescale
values, 4. filter, 5. rescale values, 6. replace channel, 7. convert to
RGB).

After playing around with this a bit more, I agree that defaulting to
filtering each layer of RGB images is the way to go (in most cases).

-Tony

Updated @adapt_rgb: https://gist.github.com/3909400
I'm inclined to move away from the "one decorator to rule them all"
approach, and use separate @rgb_channel_filter (or @each_channel, or
whatever), @lightness_filter, etc.




>
> Am 19.10.2012 um 23:40 schrieb Stéfan van der Walt <stefan at sun.ac.za>:
>
> > On Fri, Oct 19, 2012 at 2:31 PM, Tony Yu <tsyu80 at gmail.com> wrote:
> >> I agree that documenting the extra filter parameter could be
> problematic. I
> >> guess that the decorator could inject that into the filter's docstring,
> but
> >> that's probably too magical. We could make it a hidden parameter and
> provide
> >> utility functions that just change the default behavior.
> >
> > I'd prefer for the parameter to be visible and documented for each
> > function that chooses to use it.  I think we should just make
> > application to each individual layer the default, since that's the
> > most obvious generalization to RGB.
> >
> > Stéfan
> >
> > --
> >
> >
>
> --
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-image/attachments/20121020/f5f0e7e6/attachment.html>