processing map from video source

Hi,

I have my M2P video footage from video inspection + Litchi flight log which is full readable by your log viewer tool. Do you have any hints on how to process this footage in order to have orthofotomap & 3D Model made by you?

I could extract image frames from footage and then using exiftool assign specific EXIF values from flight log, but maybe you have better (easier and simpler) workflow for achieving this?

Cheers

Jakub

Jakub 30 Desember, 2020 07:56

Komentar resmi

We do not recommend doing that. Video has a lot of rolling shutter effects in it that will cause problems during processing.

Zane 30 Desember, 2020 16:48

Harap masuk untuk memberikan komentar.

4 komentar

Clear, thx!

But.... Would this be an issue if I don't have fast moving objects in my frame?

I'm flying relatively slow (10m/s) and my shutter was set to 1/240s.

Jakub 30 Desember, 2020 20:39 0 suara

Yes because your aircraft is the fast moving object. Sometimes. That's the problem. Also, rolling shutter is worse when taking video.

Also, how are you going to control overlap?

If you are going to make a map it is best to record data specifically for that.

Zane 30 Desember, 2020 22:29 0 suara

OK, clear. I'll give it a try and compare two outputs of the same area - one with still images and the other with frames extracted from video footage.

Regarding your question about overlap - if I extract frames with 1sec interval and my drone moves at constant speed 10m/s, then I will have dX=10m which at 60m AGL will be definately enough for overlaps.

Jakub 31 Desember, 2020 16:42 0 suara