[pdal] Calculation of the difference between DSM and DTM
Raúl Casado Barbero
raul.casado at eez.csic.es
Tue Jun 25 07:30:21 PDT 2024
Hello Andrew,
yes I'm sure. In fact I have some specific nodata pixels in DTM but the
negative values in the difference cover almost the entire image.
I'm using QGIS to substract both rasters using Raster calculator and
GDAL Raster calculator but I get identical result.
El 25/06/2024 a las 15:58, Andrew Bell escribió:
> Hi Raul,
>
> Are you sure you're not seeing places where you have NODATA in one
> raster or the other?
>
> On Tue, Jun 25, 2024 at 2:09 AM Raúl Casado Barbero
> <raul.casado at eez.csic.es> wrote:
>
> Hi,
>
> thank you both. Imodified the pipeline adding origin_x, origin_y,
> width
> and height and now I get consistency inthe new rasters.
>
> The problem is that I still get negative values when I substract
> the MDT
> to MDS. Could it be due to interpolation in the calculation of
> MDT? If I
> get this raster by filtering the vegetation, could it be that values
> lower than MDS appear in areas where there is vegetation?
>
>
> Greetings.
>
> El 20/06/2024 a las 16:51, Connor Manning escribió:
> > For general information on how pipelines work:
> > https://pdal.io/en/2.7-maintenance/pipeline.html
> > And here is the information on the GDAL writer specifically, take a
> > look at origin_x and origin_y:
> >
> https://pdal.io/en/2.7-maintenance/stages/writers.gdal.html#writers-gdal
> >
> > For your pipelines, the GDAL writer is independently choosing
> its own
> > grid each time since you have not specified an origin, so the pixels
> > don't line up between the two outputs. You want to explicitly
> set the
> > same origin, via the above options, for both of your writers.gdal
> > outputs to make sure they line up.
> >
> > - Connor
> >
> > On Thu, Jun 20, 2024 at 8:36 AM Raúl Casado Barbero via pdal
> > <pdal at lists.osgeo.org> wrote:
> >> Hi Andrew and thaks for your answer.
> >>
> >> I'm pretty new with pdal just now and I'm tryingn to understand
> the logic of pipelines... What do you mean with set the origin on
> writers.gdal?
> >>
> >>
> >> Raúl.
> >>
> >> El 20/06/2024 a las 13:45, Andrew Bell escribió:
> >>
> >> Hi Raul,
> >>
> >> You should set the origin on writers.gdal in order to get the
> consistency you desire. Without this, the origin is set according
> to the first point processed, which may differ between pipelines.
> >>
> >> On Thu, Jun 20, 2024 at 4:24 AM Raúl Casado Barbero via pdal
> <pdal at lists.osgeo.org> wrote:
> >>> Hi,
> >>>
> >>> my aim is to calculate the fitovolume of a shrub vegetation
> area. To do that I calculate the DSM, classify a point cloud with
> ground (2) and low vegetation (3) and then create a DTM.
> >>>
> >>> I use 2 different pipelines (see below) to do that modified
> from this post. But I have 2 issues:
> >>>
> >>> When I calculate the difference between DSM and DTM I get
> negative values
> >>> The pixels of the DTM are horizontally displaced with respect
> to the DSM by half a pixel.
> >>>
> >>> Does anyone know why I am getting these negative values using
> this pipelines? Is anything wrong in the pipelines I`m using?
> >>>
> >>>
> >>> Raúl.
> >>>
> >>> Pipeline for DSM:
> >>>
> >>> [
> >>> "nubeRecortadaPdal_1.laz",
> >>> {
> >>> "type": "filters.range",
> >>> "limits":"returnnumber[1:1]"
> >>> },
> >>> {
> >>> "type": "writers.gdal",
> >>> "filename": "nubeRecortadaPdal_1_mds.tif",
> >>> "output_type": "idw",
> >>> "gdaldriver": "GTiff",
> >>> "resolution": 0.05,
> >>> "nodata": -9999,
> >>> "radius": 0.07,
> >>> "window_size": 50,
> >>> "data_type": "float32"
> >>> }
> >>> ]
> >>>
> >>> Pipeline for DTM:
> >>>
> >>> [
> >>> "nubeRecortadaPdal_1.laz",
> >>> {
> >>> "type":"filters.overlay",
> >>> "dimension":"Classification",
> >>> "datasource":"D:/Datos/clasificadasVector.gpkg",
> >>> "layer":"ndviRandomForest",
> >>> "column":"Id"
> >>> },
> >>> {
> >>> "type":"filters.range",
> >>> "limits":"Classification[2:2]"
> >>> },
> >>> {
> >>> "type": "writers.gdal",
> >>> "filename": "nubeRecortadaPdal_1_mdt_ndviRf.tif",
> >>> "output_type": "idw",
> >>> "gdaldriver": "GTiff",
> >>> "resolution": 0.05,
> >>> "nodata": -9999,
> >>> "radius": 0.07,
> >>> "window_size": 100,
> >>> "data_type": "float32"
> >>> }
> >>> ]
> >>>
> >>>
> >>> _______________________________________________
> >>> pdal mailing list
> >>> pdal at lists.osgeo.org
> >>> https://lists.osgeo.org/mailman/listinfo/pdal
> >>
> >>
> >> --
> >> Andrew Bell
> >> andrew.bell.ia at gmail.com
> >>
> >> --
> >> Raúl Casado Barbero
> >> Servicio de Evaluación, Restauración y Protección de
> Agrosistemas Mediterráneos
> >> Estacion Experimental del Zaidin, CSIC
> >> Calle Profesor Albareda 1, 18008 Granada, Spain
> >> Phone: +34 958181600 VolP CSIC 439186
> >>
> >> _______________________________________________
> >> pdal mailing list
> >> pdal at lists.osgeo.org
> >> https://lists.osgeo.org/mailman/listinfo/pdal
>
> --
> Raúl Casado Barbero
> Servicio de Evaluación, Restauración y Protección de Agrosistemas
> Mediterráneos
> Estacion Experimental del Zaidin, CSIC
> Calle Profesor Albareda 1, 18008 Granada, Spain
> Phone: +34 958181600 VolP CSIC 439186
>
>
>
> --
> Andrew Bell
> andrew.bell.ia at gmail.com
--
Raúl Casado Barbero
Servicio de Evaluación, Restauración y Protección de Agrosistemas Mediterráneos
Estacion Experimental del Zaidin, CSIC
Calle Profesor Albareda 1, 18008 Granada, Spain
Phone: +34 958181600 VolP CSIC 439186
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/pdal/attachments/20240625/bf2b4f99/attachment-0001.htm>
More information about the pdal
mailing list