Forestry canopy top mapping

[Crunchie]

Hi everyone

I am trying out a Agisoft for the first time as I would like to calculate forestry canopy top levels. We currently use a medium format camera (80mm which included GPS and Roll, Pitch and Yaw data). I have had moderate success with some trial runs done with 25cm GSD imagery and 70/60 forward and side overlaps (moderate success = 0.5m to 5m variations on ground and tree canopy top levels when compared to LiDAR data of the area). I would appreciate it if any of you could point me in the directions of a custom tutorial or advise on settings/workflows to use in order to improve accuracies.

Thanking you in advance.

Crunchie

[Crunchie]

Thanks Nadir, I don't have the luxury of doing a full photogrammetry solution hence the reason for trying agisoft....

Any other pointers that may assist, like increased overlap, lower resolution?

[Crunchie]

Also, added to that I am not expecting pin point accuracy at all, mainly a consistent average of the canopy to within 1-3m over the target as opposed to the big variations (0.5m to 8m) I am currently getting.

Thanks

[Matt]

I have been playing with canopy top mapping for a few years and wind conditions have been the key for me.  I pick days with winds of less than 5 knots and good clear skys.  Been getting average accuracies averaging +/- 0.5 metres at 10 cm GSD from 600 m flying height with high precision ground control. The main problem for me has been getting lidar that is properly  georeferenced to compare with. 

[Franco]

I've created nice models of short vegetation (sagebrush, bitterbrush, willows, etc.) and the one thing I've learned to try to avoid is wind that moves the veg. That's a model killer. Sometimes it's just not possible to avoid, and in such a case I find taking more photos with 80-90% overlap helps. Not all points will be retained during optimization because of poor base to height ratio, but it at least allows image alignment whereas often 60% overlap in such conditions fails.