I'm currently doing a historical research project on a hospital that no longer exists. I've got a detailed site plan that I would like to geo-reference so I can digitize it. However, the site plan is broken up into a bunch of jpegs taken with a phone camera from the original paper copy. It is not properly scanned as a single file.
While they are perfectly readable, each photo is taken at a slightly different angle or height, as well as overlap. This poses a problem where if I just sew the images together in photoshop before geo-referencing, I will have a single plan with potentially wildly differing scales.
My only idea right now is trying to geo-reference each jpeg individually to smooth out the differences before mosaicing them together, but is there a better way to solve this problem? Ideally I would like to capture the precision of the original plan.