Hi Pete,
Thank you. That was really useful

Roberto

2014-09-25 9:18 GMT+02:00 Pete Bunting <[email protected]>:
> Hi Roberto,
>
> Yes, that is correct. spdmccgrd and spdpmfgrd only assign points to ground
> and spdinterp and spddefgrd just consider ground / non-ground returns.
>
> Cheers, Pete
>
>
> ****************************************************
> * Dr Pete Bunting
> * Senior Lecturer in Remote Sensing
> * Earth Observation and Ecosystem Dynamics Group
> * Department of Geography and Earth Sciences
> * Aberystwyth University
> * Aberystwyth
> * Ceredigion
> * SY23 3DB
> * UK
> *
> * Ph: +44 (0) 1970 622615
> * Mob: +44 (0) 7917 842743
> * Email: [email protected]
> * ORCID: http://orcid.org/0000-0002-7435-0148
> ****************************************************
>
> On 24 Sep 2014, at 22:12, Roberto Antolín <[email protected]> wrote:
>
> Hi Pete,
> Are you saying that the only classification that pmf and mcc assign is
> "ground"? That explains lots of things :)
>
> So, I guess that the data I have with "high vegetation" class was
> already classified before applying spdpmfgrd and spdmccgrd, right?
>
> Then, what happens with spddefheight and spdinterp? I suppose they use
> ground and no-ground points in order to define the reference surface
> for height definition and the DSM, DTM and CHM, respectively.
>
> Cheers,
> Roberto
>
> 2014-09-24 19:45 GMT+02:00 Pete Bunting <[email protected]>:
>
> Hi Roberto,
>
> I am not sure what is going on there. There haven’t been any chances to that
> code over the period you talk about. However, I think the difference is more
> likely to be with your input data. The spdmccgrd and spdpmfgrd commands do
> not try to classify vegetation or any other classes only ground but if
> returns have a classification before they are entered into spdmccgrd or
> spdpmfgrd then only the points which are classified at some point with the
> processing as ground are changed. In the final classification not all point
> may end up as ground, if they are only ground at an intermediate stage then
> those points will be assigned as unclassified. Therefore, you will commonly
> see quite a few unclassified returns near the ground.
>
> If you want to test it or to ‘clean’ your data before using spdlib to
> classify it then you can use the spdclearclass command to assign all returns
> to the class ‘unclassified’.
>
> I hope that helps, let me know if it doesn’t make sense or this is not what
> you are seeing.
>
> Cheers, Pete
>
>
> On 24 Sep 2014, at 16:57, Roberto Antolín <[email protected]> wrote:
>
> Hello all,
> I have recently noticed that PMF and MCC algorithms do not classify
> vegetation properly. The output spd files has two (some times three)
> different classes: "ground" and "unclassified" ("low vegetation" in
> some rare cases). None of them are "vegetation"
>
> Since I have not updated my spdlib version for a long time, I checked
> the files I processed months ago. As far as I know, this issue is
> going on since May.
>
> Any hint?
>
> --
> Roberto
>
> ------------------------------------------------------------------------------
> Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
> Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
> Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
> Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
> http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
> _______________________________________________
> spdlib-develop mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/spdlib-develop
>
>
>
>
>
> --
> Roberto
>
>



-- 
Roberto

------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
_______________________________________________
spdlib-develop mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spdlib-develop

Reply via email to