GCPs load, but in the wrong place

I believe this must be a formatting issue. I’ve only ever had partial success with GCPs. Please let me know what I’m missing in my .txt file.


GCP Interface issue|690x392

1 Like


Sorry you’re having a bit of trouble with your GCPs.

Are you able to provide the text for them here as well?

1 Like

42.71569216203569	-72.44409890359766	65.16426547619042	a
42.71596591772522	-72.44418250470855	64.03324834437083	b
42.71562223124421	-72.44383059701983	67.65279009900988	c
42.71548601562463	-72.4435629668339	66.79212624584716	d
42.71597845746121	-72.4435404530382	67.73976805721098	e
42.71561768033766	-72.44335798582148	68.76945980707391	f
42.715503582862304	-72.44309400641902	69.75195741998057	g
42.715212639877876	-72.44323520583497	71.45065346534656	h
42.71521555789183	-72.44297180936722	79.03381672131147	i
42.71546956301539	-72.4426360418615	75.73504892307693	j
42.71514302364149	-72.44259130646049	77.74934078947369	k
1 Like

These are coordinates I gathered using an Emlid Reach rs+. The original export was a .csv file (which ODM does not read). So I manually created the above .txt file and tried to use the proper formatting.

My other attempt with GCPs used the same format but with a different issue: the GCPs that loaded were in the correct spot, however only 3/6 were pulled from the list and rendered on the map.

1 Like

Thanks for providing that!

We’ll get this one sorted and figure out what’s wrong, then I’ll try to help you with the one that dropped entries :slight_smile:

Does this location look correct? Let me know and I’ll post my corrected version.

If so, it was an issue with the state of your GCP data. It wasn’t complete/well-formed to be a gcp_list.txt yet, but was sufficient to be formatted as a CSV file for import/resume. To be a gcp_list.txt it will need the image_x, image_y, and image_name columns/values to be valid, which this does not have yet.

Check the docs here:

Yes, those are the correct locations.

How do I get the image_x,y, and name values before generating the GCP targets on the map? Don’t those values get generated only after each GCP has been linked to 5+images? Or am I missing a step?

I thought the workflow went:

Collect GCP field data
Collect drone images
Convert GCP export (csv, shapefile, GeoJSON) into “gcp_list.txt” (without image_x,y,name values)
load gcp_list.txt into GCP Interface box1
load drone images into GCP Interface box2
Manually ‘tie’ GCP targets on map to visible GCPs in drone images (5+images per GCP)
“Export File” from GCP Interface (now includes image_x,y,name values of manually selected GCPs).

Then my assumption (although I have yet to reach this step) is:

upload drone images AND newly exported GCP file into a new project (from dashboard) before setting parameters

I could use some help clarifying this workflow. :sweat_smile:

1 Like
GCP Label,Y,X,Z
a, 42.71569216203569, -72.44409890359766, 65.16426547619042
b, 42.71596591772522, -72.44418250470855, 64.03324834437083
c, 42.71562223124421, -72.44383059701983, 67.65279009900988
d, 42.7154860156246, -72.4435629668339, 66.79212624584716
e, 42.71597845746121, -72.4435404530382, 67.73976805721098
f, 42.71561768033766, -72.44335798582148, 68.76945980707391
g, 42.71550358286230, -72.44309400641902, 69.75195741998057
h, 42.715212639877876, -72.44323520583497, 71.45065346534656
i, 42.71521555789183, -72.44297180936722, 79.03381672131147
j, 42.71546956301539, -72.4426360418615, 75.73504892307693
k, 42.71514302364149, -72.44259130646049, 77.74934078947369

So, with the field-collected GCP/GPS data, you only really have enough to form a proper gcp.csv, the format/structure of which is clarified in the docs:

From there, you’d load it into POSM GCPI or GCP Editor Pro as a CSV to then tag the images and the image x/y locations which are mandatory for a properly formed gcp_list.txt

So, not terribly different from how you understood the workflow, but just that you have to start with the gcp.csv when you start from the GPS data.

A post was split to a new topic: How to put gcp_list.txt in Project File