CHECK POINTS Versus GCP

When is it best to use CHECK POINTS rather than GCP?
I recently watched a webinar about using GCP to set the scale and Z orientation for a dataset with no georegence tags. I am wondering if the same could have been accomplished with check points, yes, no and why?

Check point will not help you. They are used to give you accuracy statistics fully indepentent on the processing. Check point are not used in the optimization only for thep accuracy reporting.
Pix4Dmapper 1.5 will include more easy ways to set the scale and orientation. For now, you should used 3D GCPs

Christoph, I don’t see the value of checkpoints then, I guess I am not understanding their use.
What do you mean they are used to give me an accuracy statistic?

You should use Check Points for a known location so that you can see the amount of error in the project, shown in the quality report and when you select the point in the project. This is very helpful so that you can be confident in your results or know that you need to add to it for more accuracy.

I think we need a video at some point as an example. For some reason I am having a very hard time understanding this feature.

Dear DDS CAD,

The check points are not taken into account for processing and calibrating the model.
They are used only to assess the accuracy of the project.

In order to scale, locate and orientate the project and improve the relative and absolute accuracy of the model, you should mark GCPs in the project.
A minimum number of three GCPs is needed, but we recommend between 5-10, well distributed over the area, depending on the terrain, the difficulty of the area and the height differences.
For more information about GCPs: https://support.pix4d.com/hc/en-us/articles/202557489
If you have abundant GCPs, then you can mark them in the project as Check Points in order to assess the accuracy of the project.

For more information about the Quality Control of a project: https://support.pix4d.com/hc/en-us/articles/205148245-

“The check points are not taken into account for processing and calibrating the model.
They are used only to assess the accuracy of the project.” … by comparing its true predetermined geo referenced locations (XYZ) (values from an RTK GPS) with the position of a calculated resulting point.

I added the last part of the sentence which is my understanding based on Oliver’s last seminar on project accuracy.

Hi there - I understand the purpose of checkpoints - however I am having trouble getting the quality report to generate the error information for the check points I have included. 

I have imported checkpoints along with GCPS

I have changed the Type from GCP to Check Point to ensure they don’t get included in processing, then Apply

The points seem to sit on the resulting surface - but I was expecting to see the error information for the check points listed below the error information for the GCP’s in the quality report - as I have seen in various webinars.

Am I missing something?

Any help would be appreciated. Thx Jon

 

I am referring to how to generate XYZ error information for check points in the Quality Report.
Thanks.

Dear John,

In order to compute the difference between the initial and the computed coordinates of the check points, you need to mark them on the images, reoptimize and regenerate the quality report as you would do with normal GCPs. The only difference is that the check points will not participate in the georeference of the model. 

You can watch this video for more information regarding the different types of tie points.

Best regards,

1 Like