ODM Parameter Suggestions for clearer orthophoto image

[israelbar]

@darkl1ght

I’m suspicious on the tiles, what if you download the .tif orthophoto file and visualize in qgis?

I’m also curious to see what happens if you change texturing-data-term to area.

[darkl1ght]

Hi @MikeF,

Drone is DJI PHANTOM 4 PRO
and sensor details are present here: https://www.dji.com/phantom-4-pro/info#specs

Any details you could provide for camera lens correction flag? after getting above details.
Thanks

[Saijin_Naib]

Phantom 4 Pro Rolling shutter effect

CMOS sensor with typical CMOS readout (rolling shutter), but with a mechanical leaf shutter available down to 1/2000s exposure to help compensate for some of that distortion.

So, without ODM having rolling shutter compensation, there might be bowling/distortion, even with the Brown-Conrady camera model, especially if you acquired the photos while moving quickly.

[MikeF]

Camera Types

The type tells ODM which mathematical model to use to correct lens distortion in the photos.

Saijin describes the rolling shutter issues very well the previous post. (Thanks to @Saijin_Naib who always provides keen insights)
M

[JulianW]

It may be worth mentioning that the mechanical shutter on the P4P has to be enabled as well. I don’t recall if it’s on by default though.

[Saijin_Naib]

Thanks for the support, Mike!

If our current fund-raiser goes well, I think the next area I’d like to tackle with a fund-raising effort is a rolling-shutter compensation model. We all keep running across scenarios where it would help, and given the proliferation of CMOS sensor platforms, I think it will remain relevant for a long time yet. All of my data suffer from it (captured in YOLO-Mode on my Solo [35mph ]), so I feel the need, for sure.

Julian makes a great point about implementation… I have no idea if that is on/off by default, so please check your setup.

[kikislater]

@pierotofy : What is the current status in ODM of DJI altitude/height correction ? Are they corrected as needed or stay as this ?

[darkl1ght]

HI @israelbar,
Will try this visualization in Qgis

And re-process with texturing-data-term to area instead of gmi.
Will keep you posted.

[israelbar]

Hey @darkl1ght, I processed one of my datasets changing texturing-data-term from gmi to area and it did not make it better. It’s different but not better.

But visualizing in qgis improve the quality of the orthophoto. I guess the tiles are not making honor of the final result of the tiff file.

Saludos

[darkl1ght]

Hi @israelbar,

Thanks for this info. Even I observed the same. (Now I know I am getting issues because of the shuttering effect of DJI Phantom 4 Pro camera)

@israelbar

It’s different but not better.

Yet to visualize in QGIS.

Update: Thanks for the suggestion. So, I visualized in QGIS and its around 70% better compared to WEBODM.

Questions:

• Why does this happen?
• How to use the original quality of the orthophoto.tif file data?

Thanks

[smathermather]

I have been seeing scattered reports of this since the introduction of cloud optimized geotiffs to WebODM. @darkl1ght and @israelbar – can you document this difference with screen shots?

[darkl1ght]

Hi @smathermather,
what should be the document format? Is mere comparison enough?
and can those screenshots be annotated with drawings?

Thanks

[smathermather]

That’d be perfect.

[darkl1ght]

Hi,

So this is the document link. Let me know if someone wants to contribute more data or need any modifications.

Thanks

[israelbar]

@smathermather

can you document this difference with screen shots?

I found same results as @darkl1ght, visualization is slightly improved when using in qgis.

@darkl1ght these straight lines in your orthophoto are really challenging, but you are nearly there…

In my short experience using ODM, I have seen these straight lines improved with the 3D model quality. It seems that the more accurate 3D model, the more accurate orthophoto, especially for these straight lines.

Saludos

[israelbar]

@darkl1ght Does the 3d model shows this same issue?

[darkl1ght]

@israelbar cant figure that out from the 3D-model.

This is the original image from dataset

[israelbar]

As I expected, there is a lack of points in the area where the orthophoto is deformed.

So, I would suggest to plan your flight as for a very accurate 3D model, I think it is a matter of designing an adequate flight plan for this project to work in OpenDroneMap.
Parts of point cloud not showing up on textured model - #10 by JulianW

I’m sure you will figure out how to create an orthophoto with perfect straight lines.

[darkl1ght]

Thanks, will take care of that next time