Path: news.mathworks.com!not-for-mail
From: "Peter Bone" <peterbone@hotmail.com>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Merge line segments
Date: Tue, 12 Feb 2013 16:07:18 +0000 (UTC)
Organization: The MathWorks, Inc.
Lines: 13
Message-ID: <kfdpbm$ke8$1@newscl01ah.mathworks.com>
References: <kfb25v$opl$1@newscl01ah.mathworks.com> <kfb5tu$b93$1@newscl01ah.mathworks.com> <kfb6hv$dqb$1@newscl01ah.mathworks.com> <kfbaad$sur$1@newscl01ah.mathworks.com>
Reply-To: "Peter Bone" <peterbone@hotmail.com>
NNTP-Posting-Host: www-04-blr.mathworks.com
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: newscl01ah.mathworks.com 1360685238 20936 172.30.248.35 (12 Feb 2013 16:07:18 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Tue, 12 Feb 2013 16:07:18 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 870367
Xref: news.mathworks.com comp.soft-sys.matlab:788824

"Bruno Luong" <b.luong@fogale.findmycountry> wrote in message <kfbaad$sur$1@newscl01ah.mathworks.com>...
> "Peter Bone" <peterbone@hotmail.com> wrote in message <kfb6hv$dqb$1@newscl01ah.mathworks.com>...
> > 
> > Thanks. That's a nice method for minimising code and works fine, but it's around 6 times slower than mine on average.
> 
> Next suggestion:
> If you have a bunch of quadri-points to be processed, then vectorize your code rather than calling the function sequentially.
> 
> Bruno

Thanks, but I don't think that's possible. They have to be processed sequentially because the next line segment to be processed could merge with a line segment that has been previously merged.

I was hoping for something more algorithmic because I will be coding this in C++ eventually anyway. I was perhaps thinking there could be something more efficient by not having to compute the square distance between each each pair of points. Perhaps that isn't possible.