Re: Focus stlyes for non-standard shapes

The main point for me was whether for irregular shapes of any kind, approximating the measurement by taking the longer side of the box enclosing the shape and the thickness of the irregular focus border/shadow/outline within it would be a permissible approximation. I think it would be. (I doubt it will be feasible to automate anything based on screen captures).
Cheers,
Detlev

> Am 17.10.2019 um 00:01 schrieb Alastair Campbell <acampbell@nomensa.com>:
> 
> > but would it not be far easier to …
>  
> Yes, in practice that would be a much easier method.
>  
> I wanted to tackle something complex, with a gradient, and show there is a way to measure it. That could work for a spiral shape that has no straight lines.
>  
> It is a huge faff, but it would be a very rare instance as every focus style I’ve tested in the wild is either rectangular (~99%) or circular. (In fact, I think the circles were rectangles with border radius.)
>  
> If this sort of testing did somehow become common, that method (basically pixel-counting the difference) seems like something you could automate with screen captures [1].
>  
> Cheers,
>  
> -Alastair
>  
> 1] https://www.pyimagesearch.com/2014/09/15/python-compare-two-images/ <https://www.pyimagesearch.com/2014/09/15/python-compare-two-images/>
> Tool to give a score of the difference between two images, not ready for this purpose, but would be possible to create a front-end for testing screencaptures.