Anybody else have this problem or is this just my camera??

Were you ascending/descending at the moment this was taken or was the camera still?    What was your shutter speed?

Cool shot by the way, like what you've done with it.

I've seen it on mine as well.   I was shooting a hotel and beach.  In the upper/right of the video, there was a wave.

The camera was pointing downward at the time, at about the same angle as in your shot above.  The wave showed up when the Inspire was in a stationary hover -- and whether the camera was still or panning.  We shot the same scene with the camera pointed a less of an angle and the wave disappeared.

I was in Auto when I shot, 4K - 30fps.

I had seen reports of this early on when the Inspires first came out, but haven't heard much since then.

It seem to have something to do with the downward angle of the camera, but I've seen no explanation.

jimhare Posted at 2015-3-10 12:22
Were you ascending/descending at the moment this was taken or was the camera still?    What was your ...

Hi Jim,
Sorry about the delay but I had Notifications turned off.
I was stationary at the time. Here's a screen shot of my settings... I'm not the greatest photographer!!

MJDMike Posted at 2015-3-10 17:46
I've seen it on mine as well.   I was shooting a hotel and beach.  In the upper/right of the video,  ...

Mike,
Sorry for the delay ... had notifications turned off!!
I think the I1 has a little fisheye in it ... so if the horizon is not in the middle this will show up.
I'm looking for a post where someone has posted a fix for this as a Photoshop plugin.
Let me know if you have any ideas.

air.raid Posted at 2015-3-22 01:05
Hi Jim,
Sorry about the delay but I had Notifications turned off.
I was stationary at the time. Her ...

Hey.  Those settings should have given you a rock solid photo so seems like nothing you could have done differently, especially if you were stationary.

Be interesting to see what others say about it.

I do not work for DJI or Sony (who made the camera).
Still, I know little about CCDs made of 3x3 panels since many digital aerial cameras used in photogrammetry were designed in a similar way and I had to understand the kind of distortions they had, in order to support correction formulas in post processing applications.
My understanding is this:
- lens of the camera produce a radial distortion (Conrady coefficients correction); each CCD would receive a shifted radial distortion from the lens (and possibly a small perspective distotion)
- digital images are being captured electronically either through a "Rolling shutter" or "Global shutter" method; CCDs often use the "Global shutter" technique, but we still have 9 of them and as much as we want to believe, they are never in perfect sync of capturing the "same" snapshot; try recording 4096x2160 24P video and rotate the I1 while the camera is tilted at about 30 degrees from perfect horizontal; extract all frames from your video and you will notice many consecutive frames having only the upper regions changing (due to perspective distortion combined with the small out-of-sync capturing of the individual CCDs); if these CCDs would use "Rolling shutter" the effect would most probably be worse, but visually different
- each CCD also requires an individual resampling / scaling adjustment; moreover, these individually captured panels of pixels need to be "stitched" together to form a single digital frame
- in order to construct an entire digital frame, the camera does internally apply some corrections and processes; some professional cameras, similarly to other measurement equipment, require and allow the possibility to re-calibrate themselves, by recomputing the coefficients used in the applied formulas; high differences in temperature, certain shocks / vibrations and other events may require re-calibration of a camera
- distortions can also be measured and partially removed during the post processing of the captured imagery; when the source / type of the distortion is known, before or even after capturing the imagery, the camera distortions can be measured and then the data (imagery) can be corrected, unless the distortion aberration is within acceptable range
Hoping that I haven't bored you all, we should probably do some tests and measure the distortions our I1 cameras have. As long as we follow the same procedure and carefully perform the steps, we should be getting almost the same distortion results. Only then we will really know how good is this camera for photogrammetry and also if one of us has a defective unit.
Best regards,
Bogdan

If you do a search over on RC Forums their are some lens profiles for DNG and JPEG for both Lightroom and Photochop that will get rid of this lens distortion.

RC Group

[url=http://www.rcgroups.com/forums/showthread.php?

t=2280074&page=1285#post30869901]DNG[/url]

The glass on the Inspire is rectilinear(ish) for a 94degree FPV lens but it's not a high quality wide angle prime so expect some straight line distortion. What you are seeing is normal.
Edit..... My second link didn't work for some reason.... I have no idea why it didn't embed sorry.

I saw this on some images from a flight this morning. This looks like a 'seagull' lens distortion and is very common in many lenses. This is why most photo editing software have lens correction for many specific lenses as each lens has its own errors (pincushion, barrel, seagull etc) or a combination. Thus all my Nikon glass have profiles, my Leicas and I don't know about my Fujis because they don't have distortion. I see that Lightroom has a profile set for DJI but only for the Phantom and not the Inspire 1 yet. If you care about this then Photoshop is your friend. Personally I do not think it detracts from your picture especially as you have added a strong vignette.

ian Posted at 2015-3-23 02:26
I saw this on some images from a flight this morning. This looks like a 'seagull' lens distortion an ...

Adobe Camera Raw 8.8 now has the profile for the Inspire 1 but at the moment this is only available for Photoshop and not Lightroom. It makes a big difference and sorts out a barrel distortion very well.