• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

Error with 'DETECTFEATURES'

ibanez737

ibanez737

Messages
153
Country
unitedstates
I'm getting an error saying "DETECTFEATURES should only be used after the SPLITGRID step" while trying to create vegetation from a raster. As the error suggests, I'm not using the SPLITGRID step, but my config file is set up almost exactly like the one in the tutorial video HERE and that doesn't show any errors. Here is a text of my config file:

IMPORTGDAL|E:\FS\FSX\Projects\Data\Vegetation.tif
#
DETECTFEATURES|FTYPE=RASTER|Filter_Histogram|OBJ|Forest
#
EXPORTSHP|*|SF_Vegetation

Anyone know what's going on?

Thanks,
Scott
 
is your tiff edited and or properly geo referenced?
does your tiff has the proper alpha channel needed for Histogram to work properly?
if not use the other method Arno provided (non Histogram based)




Chris
 
Last edited:
Hey, thanks for the reply Chris.

is your tiff edited and or properly geo referenced?
Click to expand...

Yes, it's georeferenced with geographic projection and using WGS84 datum. This shouldn't matter for this error though.

does your tiff has the proper alpha channel needed for Histogram to work properly?
Click to expand...

I believe you're talking about the 4th 'near infrared' band, and to be honest, I'm not sure if my imagery has it or not. I am using IR imagery though, with the vegetation appearing red due to the false coloration which is giving me pretty good results with using the histogram method. I couldn't get the 'other' method to detect any veg at all (can't remember what the other method is called either, lol). But again, I don't think this would be causing the error to appear in the 'error' box at the bottom of the GUI...

Cheers,
Scott
 
lol, im still confused with the two names myself but aware of the differences
honestly i don't have much mileage using veg detect feature with ScenProc
i can can go as far as point the simple and obvious :o

wait for Arno to pick this one up; cause i have no clue further


regards


Chris
 
Hi Scott,

That requirement to have the splitgrid step before the feature detection is new, that's why it's not in the tutorial (yet).

Recently I found out that the performance is much better when processing smaller areas at once. Especially if I combine that with using multiple threads. Before only a few threads were used, now it increased that amount. However processing big photos with so many threads would choke almost any pc and use too much memory, so that's why I introduced the requirement to have the splitgrid before.
 
Well OK, now I'm getting another error when trying to run this project. I added the SPLITGRID|AGN entry before the DETECTFEATURES step and ran the file, and everything seemed to be going smoothly for about 50 minutes before this error popped up:

rim26.jpg


I'm getting the same error when running on an identical configuration on a different computer, so I really have no idea what it's talking about when it says "other memory is corrupt". No output files resulted. Anyone know what's going on?

Thanks,
Scott
 
Hi Scott,

I saw the bug you reported, it seems to happen when saving to shp.

How many features were created?
 
Running it again, this time I'll post the log, should have done that yesterday. If it only happens when exporting to SHP, do you think it would work better exporting to KMZ or some other form of vector? Would the next step (creation of AGN) turn out any differently?

Cheers,
Scott
 
Here's the event log:

10:50 AM IMPORTGDAL Information Reading GeoTiff file E:\FS\FSX\Projects\NCA\PR Areas\SanFrancisco\Data\IR Vegetation\Export\Vegetation.tif
10:50 AM IMPORTGDAL Information Read 1 features from file
10:50 AM SPLITGRID Information Splitting features in 0.0146484375 x 0.010986328125 grid...
10:50 AM SPLITGRID Information Features divided in 209 cells
10:50 AM DETECTFEATURES Information Starting feature detection...
11:16 AM DETECTFEATURES Information Added 51103 features
11:16 AM EXPORTSHP Information Writing SHP file SF_Vegetation
 
Hi,

That all looks quite ok. 50000 is not an amount of features that should give big trouble.

But the error message doesn't really tell me what goes wrong. Just that something went wrong in the library that writes the shapefile. So I guess the only way to fix it would be to try if I can reproduce it.

About saving to other formats, at the moment scenproc can't do that. It's on the wishlist to use OGR to save to other formats as well.
 
OK, I'm uploading my files now so you can try to reproduce the error; I'll send you an email with the link once it's complete.

I tried re-installing the Access Database x64, but that didn't change the result (not sure if it makes a difference for the new version anyways). And I'm using the 64 bit version of Scenproc on my 64 bit system.

Cheers,
Scott
 
Hi Scott,

Thanks, I got your email and have a look.

Those access drivers are only needed for the IMPORTSHP step that will be deprecated soon.
 
Hi Scott,

I know what goes wrong. I'll try to fix it tonight. The problem is that saving raster features to shapefile goes wrong. So for now you could add a filter OBJ=FOREST and then it will work.
 
Hi,

I have fixed the issue now. So in the release of tomorrow it should all be fine again.
 
Thanks for reporting the issue, I hadn't found it myself yet.
 
You must log in or register to reply here.
Back
Top