<div dir="ltr">Team,<div><br></div><div>I've opened an issue here and will continue to investigate. Any insight is much appreciated.</div><div><br></div><div><a href="https://github.com/OpenDroneMap/OpenDroneMap/issues/350">https://github.com/OpenDroneMap/OpenDroneMap/issues/350</a><br></div><div><br></div><div>Thanks,<br>Dennis</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jul 28, 2016 at 1:24 PM, Dakota Benjamin <span dir="ltr"><<a href="mailto:dmb2@clevelandmetroparks.com" target="_blank">dmb2@clevelandmetroparks.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
<div style="font-size:12pt;color:#000000;background-color:#ffffff;font-family:Calibri,Arial,Helvetica,sans-serif">
<p>So I don't see in the exif where AGL is but it looks to me like the GPS Altitude attribute is AMSL, and that is the line taken from exif to determine Z. In fact, the extract_utm command parses from a jhead output to extract the line containing the string <span style="color:rgb(24,54,145);font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:12px;line-height:20px;white-space:pre-wrap">"</span><span style="color:rgb(24,54,145);font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:12px;line-height:20px;white-space:pre-wrap">GPSAltitude"
</span><span style="font-size:12pt">from the jhead output.</span></p><span class="">
<p><span style="font-size:12pt"><br>
</span></p>
<p><span style="font-size:12pt"><br>
</span></p>
<div>
<div style="font-size:12pt;color:#000000;background-color:#ffffff;font-family:Calibri,Arial,Helvetica,sans-serif">
<p></p>
<div>
<p class="MsoNormal"><img alt="http://sig.cmparks.net/cmp-ms-90x122.png" align="left" height="122" hspace="12" width="90" src="http://sig.cmparks.net/cmp-ms-90x122.png"><a name="m_-5910325700573921582__MailAutoSig"><span style="font-size:14.0pt;font-family:"Arial","sans-serif";color:#006c56">Dakota
Benjamin<span></span><br>
</span></a><span><span style="font-family:"Arial","sans-serif";color:#006c56">Geospatial Developer</span></span><span><span style="font-size:9.0pt;font-family:"Arial","sans-serif";color:#006c56"><br>
</span></span><span><span style="font-size:9.0pt;font-family:"Arial","sans-serif""><a href="http://www.clemetparks.com/" target="_blank">clevelandmetroparks.com</a></span></span></p>
</div>
<p><a title="Ctrl+Click or tap to follow the link" href="http://www.clemetparks.com/" target="_blank"></a></p>
<p></p>
</div>
</div>
</span></div>
<hr style="display:inline-block;width:98%">
<div dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>From:</b> Khyale Nascimento <<a href="mailto:khyale@hotmail.com" target="_blank">khyale@hotmail.com</a>><br>
<b>Sent:</b> Thursday, July 28, 2016 1:29:06 PM<br>
<b>To:</b> Dakota Benjamin; <a href="mailto:opendronemap-users@lists.osgeo.org" target="_blank">opendronemap-users@lists.osgeo.org</a><br>
<b>Subject:</b> RE: [OpenDroneMap-users] How to troubleshoot orthophoto</font>
<div> </div>
</div><div><div class="h5">
<div>
<div dir="ltr">The exif data is attached. I will check the point cloud.<br>
<br>
<div>
<hr>
From: <a href="mailto:dmb2@clevelandmetroparks.com" target="_blank">dmb2@clevelandmetroparks.com</a><br>
To: <a href="mailto:khyale@hotmail.com" target="_blank">khyale@hotmail.com</a>; <a href="mailto:opendronemap-users@lists.osgeo.org" target="_blank">opendronemap-users@lists.osgeo.org</a><br>
Subject: Re: [OpenDroneMap-users] How to troubleshoot orthophoto<br>
Date: Thu, 28 Jul 2016 13:29:28 +0000<br>
<br>
<div style="font-size:12pt;color:#000000;background-color:#ffffff;font-family:Calibri,Arial,Helvetica,sans-serif">
Are you both seeing the same warping in the point cloud as well? If so then this is an OpenSfM issue. <br>
<br>
<br>
> <span style="color:rgb(33,33,33);font-size:15px">manipulation of focallength and ccd width</span><br>
<span style="color:rgb(33,33,33);font-size:15px"><br>
</span><br>
<span style="color:rgb(33,33,33);font-size:15px">From what I know OpenSfM uses ccd_width for initialization of the process and so inaccuracies shouldn't mess it up too bad. I don't remember if it is the same for focal length as well. Are you doing camera
calibration on top of this image manipulation?</span><br>
<span style="color:rgb(33,33,33);font-size:15px"><br>
</span><br>
<span style="color:rgb(33,33,33);font-size:15px">My suggestion would be to open up an issue on GitHub and document your process with as many specifics as possible. I will ping the OpenSfM dev to see if they have insights into the issue at hand. </span><br>
<span style="color:rgb(33,33,33);font-size:15px"><br>
</span><br>
<span style="color:rgb(33,33,33);font-size:15px">Khyale, you bring up an interesting point. I'll look into the code and see how it extracts altitude information. Can you upload an image with this exif or paste the output of exifread/jhead?</span><br>
<span style="color:rgb(33,33,33);font-size:15px"><br>
</span><br>
<span style="color:rgb(33,33,33);font-size:15px">Thanks all,</span><br>
<span style="color:rgb(33,33,33);font-size:15px"><br>
</span><br>
<span style="color:rgb(33,33,33);font-size:15px">Dakota</span><br>
<br>
<br>
<div>
<div style="font-size:12pt;color:#000000;background-color:#ffffff;font-family:Calibri,Arial,Helvetica,sans-serif">
<br>
<div>
<p><img alt="http://sig.cmparks.net/cmp-ms-90x122.png" align="left" height="122" hspace="12" width="90" src="http://sig.cmparks.net/cmp-ms-90x122.png"><a name="m_-5910325700573921582__MailAutoSig"><span style="font-size:14.0pt;font-family:"Arial","sans-serif";color:#006c56">Dakota
Benjamin<span></span><br>
</span></a><span><span style="font-family:"Arial","sans-serif";color:#006c56">Geospatial Developer</span></span><span><span style="font-size:9.0pt;font-family:"Arial","sans-serif";color:#006c56"><br>
</span></span><span><span style="font-size:9.0pt;font-family:"Arial","sans-serif""><a href="http://www.clemetparks.com/" target="_blank">clevelandmetroparks.com</a></span></span></p>
</div>
<a title="Ctrl+Click or tap to follow the link" href="http://www.clemetparks.com/" target="_blank"></a><br>
<br>
</div>
</div>
</div>
<hr style="display:inline-block;width:98%">
<div dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>From:</b> OpenDroneMap-users <<a href="mailto:opendronemap-users-bounces@lists.osgeo.org" target="_blank">opendronemap-users-bounces@lists.osgeo.org</a>> on behalf of Khyale Nascimento <<a href="mailto:khyale@hotmail.com" target="_blank">khyale@hotmail.com</a>><br>
<b>Sent:</b> Wednesday, July 27, 2016 9:08:19 PM<br>
<b>To:</b> <a href="mailto:opendronemap-users@lists.osgeo.org" target="_blank">opendronemap-users@lists.osgeo.org</a><br>
<b>Subject:</b> Re: [OpenDroneMap-users] How to troubleshoot orthophoto</font>
<div> </div>
</div>
<div>
<div dir="ltr">Hi,<br>
<br>
I have the same problem reported by Dennis. Large regions in my orthophoto may be skewed or distorted. I used the same set of images in pix4dmapper and it works well. I used 65 images captured from Inspire 1. The value of altitude extracted from exif data may
interfere in the reconstruction? <br>
<br>
In reconstruction, should I use the Absolute Altitude (AMSL, altitude Above Mean Sea Level) or Relative Altitude (AGL, altitude Above Ground Level)? The image's exif data gives me both. In Inspire 1, the 'Relative Altitude' is calculated from the barometer
sensor data. And I note the ODM uses the altitude AMSL in the their processes.<br>
<br>
Thanks,<br>
Khyale.<br>
<br>
<div>
<hr>
From: <a href="mailto:db@unmannedairlines.com" target="_blank">db@unmannedairlines.com</a><br>
Date: Wed, 27 Jul 2016 15:59:49 -0500<br>
To: <a href="mailto:s_mailinglist@geraschs.de" target="_blank">s_mailinglist@geraschs.de</a><br>
Subject: Re: [OpenDroneMap-users] How to troubleshoot orthophoto<br>
CC: <a href="mailto:opendronemap-users@lists.osgeo.org" target="_blank">opendronemap-users@lists.osgeo.org</a><br>
<br>
<div dir="ltr">Simon, thanks for the suggestion.
<div><br>
</div>
<div>My GPS EXIF data is all accurate from my testing. I've run it through Photoscan multiple times and have tried ODM a few times over the past several days. If you want to see the dataset you can find it here:
<div><br>
</div>
<div><a href="https://www.dropbox.com/sh/zyuxr73df4ozitu/AABixfPR-nmOHU5yL_1MOmyka?dl=0" target="_blank">https://www.dropbox.com/sh/zyuxr73df4ozitu/AABixfPR-nmOHU5yL_1MOmyka?dl=0</a><br>
</div>
<div><br>
</div>
<div>This includes the source imagery (95 files) plus the output geotiff from Photoscan. </div>
</div>
</div>
<div><br>
<div>On Wed, Jul 27, 2016 at 3:33 PM, Simon <span dir="ltr">
<<a href="mailto:s_mailinglist@geraschs.de" target="_blank">s_mailinglist@geraschs.de</a>></span> wrote:<br>
<blockquote style="border-left:1px #ccc solid;padding-left:1ex">
Hi Dennis,<br>
<br>
have you used the georeferencing option in one of the two ways possible?<br>
Could be related to wrong GPS info in the exif tags, or in the GCP-lists<br>
file...<br>
<br>
Regards, Simon<br>
<div>
<div><br>
On 07/27/2016 09:55 PM, Dennis Baldwin wrote:<br>
> Hi,<br>
><br>
> I've been working with some aerial imagery that I shot over the past<br>
> few months. Basically doing a comparison of ODM output with my<br>
> Photoscan results. Unfortunately with the latest codebase I've been<br>
> getting some pretty strange results when generating an orthophoto. You<br>
> can see what I'm talking about here:<br>
><br>
> <a href="https://www.dropbox.com/sh/extsb6o6jaoq8t2/AADRvkeINPydFVwKDaheZtzSa?dl=0" rel="noreferrer" target="_blank">
https://www.dropbox.com/sh/extsb6o6jaoq8t2/AADRvkeINPydFVwKDaheZtzSa?dl=0</a><br>
><br>
> You can see the intended results from Photoscan (after running through<br>
> gdal2tiles) here:<br>
><br>
> <a href="http://www.unmannedairlines.com/dev/tiger_stadium" rel="noreferrer" target="_blank">
http://www.unmannedairlines.com/dev/tiger_stadium</a><br>
><br>
> My question is what is the best way to go about troubleshooting these<br>
> results? I have all of the log output, but am not sure of the best<br>
> place to start. This imagery was generated with a Phantom 4 so I did<br>
> have to do some manipulation of focallength and ccd width just to get<br>
> ODM to run. Perhaps I botched something there.<br>
><br>
> Any insight of where to start would be much appreciated.<br>
><br>
> Thanks,<br>
> Dennis<br>
><br>
><br>
</div>
</div>
> _______________________________________________<br>
> OpenDroneMap-users mailing list<br>
> <a href="mailto:OpenDroneMap-users@lists.osgeo.org" target="_blank">OpenDroneMap-users@lists.osgeo.org</a><br>
> <a href="http://lists.osgeo.org/mailman/listinfo/opendronemap-users" rel="noreferrer" target="_blank">
http://lists.osgeo.org/mailman/listinfo/opendronemap-users</a><br>
<br>
_______________________________________________<br>
OpenDroneMap-users mailing list<br>
<a href="mailto:OpenDroneMap-users@lists.osgeo.org" target="_blank">OpenDroneMap-users@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/opendronemap-users" rel="noreferrer" target="_blank">http://lists.osgeo.org/mailman/listinfo/opendronemap-users</a><br>
</blockquote>
</div>
<br>
</div>
<br>
_______________________________________________ OpenDroneMap-users mailing list <a href="mailto:OpenDroneMap-users@lists.osgeo.org" target="_blank">OpenDroneMap-users@lists.osgeo.org</a> <a href="http://lists.osgeo.org/mailman/listinfo/opendronemap-users" target="_blank">http://lists.osgeo.org/mailman/listinfo/opendronemap-users</a></div>
</div>
</div>
</div>
</div>
</div>
</div></div></div>
<br>_______________________________________________<br>
OpenDroneMap-users mailing list<br>
<a href="mailto:OpenDroneMap-users@lists.osgeo.org">OpenDroneMap-users@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/opendronemap-users" rel="noreferrer" target="_blank">http://lists.osgeo.org/mailman/listinfo/opendronemap-users</a><br>
<br></blockquote></div><br></div>