[Talk-us] NHD data import question

[Alan Millar]

Well, it's been a while since I worked on it, so I can't recall exactly.  But as I recall, there were things like, say, a small pond made up of  a hundred nodes when 20 or so would suffice.  Or pretty straight lines with nodes at closely spaced intervals, where all but the first and last were completely superfluous because they were all in a line.  That sort of thing.

It is not that there is any sort of magic number which is the right number of nodes.  It is just that there were a lot more than were needed to adequately show the shape.

JOSM's "simplify way" command has some calculations it does, mostly around looking at the angles between adjacent nodes I think.  I don't remember the details, but I would recommend reading up on how it does it.


- Alan



----- Original Message -----
> From: Ben Supnik <bsupnik at xsquawkbox.net>
> To: Alan Millar <grunthos503 at yahoo.com>
> Cc: "talk-us at openstreetmap.org" <talk-us at openstreetmap.org>
> Sent: Thursday, April 28, 2011 1:58 PM
> Subject: Re: [Talk-us] NHD data import question
> 
> Hi Y'all,
> 
> What qualifies as too many nodes?  Do we have metrics for this?  I can 
> apply some kind of simplification if we can quantify what we want...
> 
> cheers
> ben
> 
> On 4/28/11 4:34 PM, Alan Millar wrote:
>>>  I've been checking some of the imported data and my general feel is 
> that it is
>>>  overdigitized.  I don't know if medium versus high reflects just 
> the quality or
>>>  the amount of digitization.  That would be something to check.  One 
> could also
>>>  just run some sort of simplification algorithm on all the data.
>> 
>> 
>>  I definitely found this to be the case for some parts of the
>>  area I did (Washington County, Oregon).
>> 
>> 
>>  I particularly remember it on
>>  areas like ponds and lakes; not so much on rivers and streams.  Most of 
> them were reasonable, but a fair
>>  number of them had way too many nodes.
>> 
>> 
>>  I used the JOSM "simplify way" quite a bit, which worked 
> perfectly in my opinion.
>> 
>>  It may be analogous to TIGER, in that NHD may have a composite dataset that 
> got created in multiple ways.  It may show up more in some places than others.
>> 
>> 
>>  - Alan
>> 
>> 
>>  _______________________________________________
>>  Talk-us mailing list
>>  Talk-us at openstreetmap.org
>>  http://lists.openstreetmap.org/listinfo/talk-us
>> 
> 
> -- 
> Scenery Home Page: http://scenery.x-plane.com/
> Scenery blog: http://www.x-plane.com/blog/
> Plugin SDK: http://www.xsquawkbox.net/xpsdk/
> X-Plane Wiki: http://wiki.x-plane.com/
> Scenery mailing list: x-plane-scenery at yahoogroups.com
> Developer mailing list: x-plane-dev at yahoogroups.com
>