use of control points (all or just some images?)

Sorry if this has been discussed earlier but I didn’t see any thing specific when I searched…

So I take a couple hundred images and want to set some control points so my components tie together correctly.
So I make a CP and set it in the problem images and some from the bulk component.
My question is:
Do I need to locate that control point in ALL images where it can bee seen, or just some and let RC work out the rest?
I’m a bit worried that RC will assume I’m trying to tell it the point isn’t visible in the non-tagged images and try to figure out a geometry to exclude the feature. But the point might be visible in hundreds of images so tagging each one will be a lot of work! Perhaps RC could add the point to all the images where it should be visible but wasn’t manually tagged as a generated CP?

Thanks in advance,

A few of the should be sufficient!

Actually, I found that very often one CP right in the center of the component/image is enough to give RC what it needs to tie them together. Then you can go on from there and see where it needs refinement.

Or a quicker way to solve it: if you can, take more images! :smiley:

Hi Jennifer Cross

No need to place CP on all images where they are present, best is if you place on at least 3 images per COMPONENT ( if multiple COMPONENTS are result of ALIGNMENT ) so at least 6 projections per CP.
From my experience its enough with 10 projections per CP. But it depend on source data quality ( if the images are not very well captured or subject is hard for photogrammetry overall ).