Have completed first mapping project on a big ash pile and delivered files to an environmental engineering client. An engineer with the client firm has asked about the meaning of -Z and +Z values in the raster files. I can view the TIF files in QGIS but can’t see -Z and +Z values he describes. What is he likely to be asking about? I’m guessing that he’s looking at a layer and the -Z is distance to the base plane and +Z is distance to the top at that vertex. If anyone can clear up my confusion, I’d appreciate it. Would like to not disappoint a new client. Thanks… Bob R>
I would like to take a look at this file. Could you please upload the raster file on our OneDrive (please include the .tif, .tfw and .prj files)?
Haven’t used OneDrive; please provide directions and I’ll upload. Or I can provide a Dropbox link to you if there’s a private message feature in the forum I haven’t found yet. I’m away from the desk for a short while but will check back here shortly. I really appreciate you jumping in. Thank you… Bob R.
OneDrive is quite straightforward: click the hyperlink and upload the files.
Teodora – I hate to be dense, but I’m not finding a hyperlink to upload to OneDrive. There’s a hyperlink in the messaging tools, but it’s to share the message thread with others. I’m afraid you’ll have to provide step-wise instructions for me to upload to your OneDrive. Otherwise I can send you a link to the Dropbox folder with all the files in it. If there’s a PM feature, I can do that easily. Please advise… R
Thanks again for your patience and reaching out. Hoping you’ll be able to help clear it up.
Please click on the word that appears in blue (the hyperlink) and then drag the files on the website that appears: hyperlink.
Done! Have to be off line for a bit. Back soon. Thx… R
PS: What time zone are you in?
Teodora – Am back and will keep an eye on forum thread for your thoughts when you have them. Regards… R
I have only received the orthomosaic, but I need to see the DSM/DTM, because they contain the elevation values.
I’ve figured out that the EXIF data is showing my MSL altitude at about -20 meters, when the captures were actually close to 140m MSL. Have bulk edited the EXIF data to 140 m and rerun it once with much improved results. Am making a few other changes and rerunning again now.
I think this will solve my problem, and I’d prefer to not waste your time on the earlier data set. I’ll return to this message thread with an update when I have current results.
It is most probably related to the geoid model, indeed. Several users have reported some problems with the EXIF data of images captured by DJI drones, please read this community post for more info: Negative Elevation Values.
Hopefully the mapping results look better now.
Getting correct MSL elevation in the photo EXIF data was the whole problem. I found an on-line freeware program that lets you edit that field easily. Since I had a bunch of photos to edit and a short time to get it all sorted out, I sprung for the fee-based version that let me edit all the photos at once. I had the GCP elevation from the rover, and I knew the AGL flight level for the camera, a little quick math provided the fixed MSL value at which all the photos were taken. Re-imported the corrected images into Mapper, and voila, everything fell into place. Mapper offered up abuncha correct (edited) photos for GCP marking, and the finished file elevations were right on.
It’s worth commenting that using GCPs doesn’t correct all the problems when the photo elevations are grossly understated. Not only do you have to locate the photos manually to mark for GCPs, but the finished file products lose absolute accuracy. A coal ash pile mapped with GCPs but with bad elevations in the photo EXIF were showing elevations varying from -10 meters to +15 meters. The actual elevations on site were about 80 meters above MSL. Put the right elevations in the photo EXIF, and everything falls into place. A thing of beauty.
Thanks for sharing the details regarding your workaround, we really appreciate it. It might have been cumbersome to solve this in Pix4Dmapper, but this is an issue related to the drone itself. Hopefully the 3D model looks better now (I wouldn’t mind seeing a screenshot of the point cloud, since you made me so curious).
Am uploading the point cloud to the one-drive link you provided if you want to look at it.
Browsed around on the one-drive screen to locate how to delete old files and didn’t find a way.
Is this not an issue for everyone flying P4P for mapping?
It looks like the issue is due to the inaccuracies of the on board GPS. Usually the elevation values arent that bad so it may be that you did not define the vertical coordinate system as ellipsoid. I general the elevation values are not dependable unless you have a RTK geotags or GCPs.