[Qgis-user] Removing artifacts in a created DEM

[Brent Wood]

Not a QGIS (or GRASS) based solution, but GMT (Generic Mapping Tools) has some excellent gridding tools, and for the same price :-)
 xyz2grd     http://gmt.soest.hawaii.edu/doc/5.1.0/xyz2grd.htmlsurface (very powerful & configureable)    http://gmt.soest.hawaii.edu/doc/5.1.0/surface.html
greenspline    http://gmt.soest.hawaii.edu/doc/5.1.0/greenspline.html
triangulate    http://gmt.soest.hawaii.edu/doc/5.1.0/triangulate.htmlsphtriangulate    http://gmt.soest.hawaii.edu/doc/5.1.0/sphtriangulate.htmlnearneighbour    http://gmt.soest.hawaii.edu/doc/5.1.0/nearneighbor.html

Surface in particular - as you increase the surface tension parameter, you will get a model which gets smoother - removing artifacts but also real spikes. Triangulation fits a surface touching each datapoint, with no smoothing, so tends to create ridges or valleys on the lines between points.
Whatever tool you use, no one algorithm fits all datasets well, which is why there are so many of them! You may need to experiment with the algorithm (command) and parameters (including grid increment size) which give the "best" result - ie: the one you like most :-)
Cheers,
  Brent Wood
      From: Nicolas Cadieux <nicolas.cadieux at archeotec.ca>
 To: qgis-user at lists.osgeo.org 
 Sent: Monday, October 19, 2015 4:19 PM
 Subject: Re: [Qgis-user] Removing artifacts in a created DEM
   
Hi,
I would go with the TIN if your points are far apart. That will not create the bumps that you see.  IDW is good but you may have to change the IDW weights from 2 (1\2, the typical gravity model) to 3 or 4 to smooth thing out.  Be careful and make sure you original values don't shift too much.  NN is very good if you have a very high density of sampling points (like LiDAR), at least in regards to the way it is implemented by GDAL.Nicolas Cadieux M.Sc.
Les Entreprises Archéotec inc. 
8548, rue Saint-Denis Montréal H2P 2H2
Téléphone: 514.381.5112  Fax: 514.381.4995
www.archeotec.caOn Oct 17, 2015 03:29, "Andreas Neumann [via OSGeo.org] " <[hidden email]> wrote:

 Hi Hernan,
 
 I think neither of the methods you used give good results for interpolating terrain data (or most other interpolations).
 
 I had good results with v.surf.rst (available in the Processing toolbox in the GRASS section). There are tons of options with the v.surf.rst - but the defaults probably work for most options. You also don't have to output all the raster results - just the interpolated surface.
 
 One of the other options is the "Interpolation plugin" which comes with QGIS (but you have to enable it in QGIS). It appears in the raster menu in as "Interpolation". It offers a TIN interpolation and IDW. I think the TIN method is the more interesting one.
 
 Hope this helps,
 Andreas
 
 On 17.10.2015 00:21, hersala wrote:
  
  Hello,
  
  I'm working with a bundle of N-S and E-W distributed DGPS point transects on gentle glacier slope. With these points I created a shp file and then I obtained several DEMs (.tif files) using the tool Raster/Analysis/Grid (Interpolation). I've used Nearest Neighbor, Runing Means and Inverse Distance with several argument values. Inverse Distance produces the best results, but there are some artifacts that persists and makes the output unusable. They give the impression as if the glacial slope has bumps, but on the contrary, it is quite gentle. 
  Can anyone give any advice in how to proceed to reduce such artifacts. 
  I'm using QGIS 2.8.2-Wien in MS-Windows. 
  Thanks in advance. 
  Hernan 
   
  
 _______________________________________________
Qgis-user mailing list
<a href="/user/SendEmail.jtp?type=node&node=5230817&i=0">[hidden email]
http://lists.osgeo.org/mailman/listinfo/qgis-user 
 
 
_______________________________________________
Qgis-user mailing list
<a href="/user/SendEmail.jtp?type=node&node=5230817&i=1">[hidden email]
http://lists.osgeo.org/mailman/listinfo/qgis-user 
 
   If you reply to this email, your message will be added to the discussion below: http://osgeo-org.1560.x6.nabble.com/Removing-artifacts-in-a-created-DEM-tp5230769p5230817.html   To start a new topic under Quantum GIS - User, email ml-node+s1560n4125267h38 at n6.nabble.com 
 To unsubscribe from Quantum GIS - User, <a href="http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4125267&code=bmljb2xhcy5jYWRpZXV4QGFyY2hlb3RlYy5jYXw0MTI1MjY3fDYzNDQ4MjQxNg==">click here.
 <a href="http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml" style="font:9px serif">NAML 
 
View this message in context: Re: Removing artifacts in a created DEM
Sent from the Quantum GIS - User mailing list archive at Nabble.com.

_______________________________________________
Qgis-user mailing list
Qgis-user at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-user

  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20151019/32b00c4a/attachment.html>