I got a better look at this.
----- Mensaje original -----
> On Thu, Dec 13, 2012 at 10:56:30PM +0000, Andrés Muñiz Piniella wrote:
> > There is currently an ISO standard being developed (ISO/CD 13083) to
> > find out what is the best method and it seems to be using canny edge
> > but using sections lines.
>
> The Canny edge detection algorithm, as I understand it,
> (a) may utilise different gradient filters
> (b) always results in one-pixel thin lines at the end
> So I'm not sure how that would work in the standard. Unless you always
> have *two* edges.
Yes, you are right, i thought it would show the top and bottom edge of the
gradient.
>
> > Would you think doing a step presentation or
> > RMS and doing sections on that to be a better solution? Or am I
> > missing something?
>
> Step gives sharp contours but it may emphasise the small features a bit
> too much. Also, it's somewhat eccentric. For general purpose, using
> the classic hypotenuse of a horizontal and vertical gradient filter is
> a safer bet. For the filters, I'd suggest the optimised Scharr gradient
>
> 3/16, 0, -3/16
> 5/8, 0, -5/8
> 3/16, 0, -3/16
>
> (and transponsed) that should lead to a more uniform angular response.
I only found sholer vertical or scholer horizontal. Not both, and no mention of
scharr.
i now realize that presentation might not be what I was looking for.
Presentation is only cosmetic and when i do a section I only see the original
data with no indication were the highlighted edges are.
i now think doing some sort of mask and later eporting the mask as a separate
image will be better.
>
------------------------------------------------------------------------------
LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
Remotely access PCs and mobile devices and provide instant support
Improve your efficiency, and focus on delivering more value-add services
Discover what IT Professionals Know. Rescue delivers
http://p.sf.net/sfu/logmein_12329d2d
_______________________________________________
Gwyddion-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gwyddion-users
