UTM Transformation Question

[kkeller]

Hello -

I’ve got a data set with accurate SCP’s and the collection was flown at 200’ and 300’. Using 200 only, I’m having an accuracy issue, roughly 2.5’ off, and I’m wondering if this is due to: a) the potential error produced by the UTM transformation, or b) a flaw in the approach to my SCP file. For reference, I saw this discussion @gjoelthiessen79 which seems related -

“I’ve been using GCP Editor Pro for GCP’s and was having GCP Error around 0.3m. It didn’t make sense to me since I was using good survey quality ground control. What I eventually discovered is that I had measured my GCP in UTM 14N NAD83 and should have converted my Lats/Longs to WGS84. I actually converted to WGS84 and then converted to UTM 14N coords and used the corresponding EPSG (32614) in GCP Editor. Error down to 0.011m.”

I’m not sure from the description what the remedy might be from a practical perspective. Should I be converting the survey control point file with foreknowledge that UTM is the target output?

I’d be happy to share the dataset and SCP’s if someone will provide a drop zone for me.

[kkeller]

Also, I’m using GCP editor Pro to set up the GCP file. ODM is interpreting the Z coordinate as meters rather than feet. My epsg 2885 is in US ft and when changing to the proj string, it shows the to_meter conversion (US). Assuming those are horizontal only, is there a way to specify z is in feet? I noticed on the format specification for GCP Editor Pro it does distinguish between (ft) and (m). I did use (ft) in my header.

Here’s the proj string but perhaps unrelated to the Z coord.

+proj=tmerc +lat_0=30 +lon_0=-84.16666666666667 +k=0.9999 +x_0=699999.9998983998 +y_0=0 +ellps=GRS80 +to_meter=0.3048006096012192 +no_defs

[kkeller]

So, easy resolution to the problem. Went to meters in the GCP file. Increased the feature and pc density to ultra. Nice looking result! 400+ images on a fairly basic machine - 1 hr 35 m.