comp.lang.idl-pvwave archive: archive » Re: Best inverse of IDLanROI::ComputeMask

Home » Public Forums » archive » Re: Best inverse of IDLanROI::ComputeMask

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: Best inverse of IDLanROI::ComputeMask [message #60820 is a reply to message #60672]

Thu, 12 June 2008 07:00

Loren Anderson
Messages: 22
Registered: August 2007

Junior Member

On Jun 11, 10:05 pm, Robbie <ret...@iinet.net.au> wrote:
> I managed to avoid the growing and shrinking by using the LEVELS=255
> keyword.
>
I'm not convinced this would solve the problem every time. When the
contour line passes exactly through the intersection of four pixels,
as it does on the diagonal, IDL has to pick which pixels are inside
and which are outside the line. In my experience, this sometimes
leads to extra pixels being selected. Does anyone have a code that
could straigten out contour lines so they are never diagonal?

> The only thing that contour misses are very irregular shapes which
> have holes in them or long protrusions which are 1 pixel thick.
>
This is a bit surprising too. Contour should pick up the holes, and
the long protrusions. Can you post examples on your website?

-Loren

Report message to a moderator

[Message index]

		Re: Best inverse of IDLanROI::ComputeMask By: Robbie on Wed, 11 June 2008 19:05
		Re: Best inverse of IDLanROI::ComputeMask By: Loren Anderson on Wed, 11 June 2008 06:16
		Re: Best inverse of IDLanROI::ComputeMask By: Loren Anderson on Thu, 12 June 2008 07:00

Previous Topic:	Re: Filling a contour
Next Topic:	BANKRUPTCY ATTORNEY

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Nov 29 05:34:37 PST 2025

Total time taken to generate the page: 0.24539 seconds