1 person found this helpful
It could be that the microscope lens has some distortion (pincushion or barrell) that throws alignment off. I don't know how you could test for (or correct) that in a microscope.
There is one way to "trick" photomerge (or auto-align, which I use mostly, but it's the same thing) into accepting images it otherwise wouldn't. Crop the adjacent image to include only the overlap, do the photomerge on those two and then put the resulting "aligned" image in with the rest. Nudge into place.
You could even do it as a two-step process, splitting the problem image into two, first the overlapping part and then the non-overlapping.
Make sure to lock the base (cropped) image so it doesn't get transformed.
I tried auto-align to see if maybe that trick worked differently from the photomerge and it still wasn't perfect. It left the unmatched picture in the middle of the canvas over another image that was aligned correctly. I hate to admit that this is the first time I am trying to use photoshop so I don't know how to try the crop and nudge idea you suggested.
Ah, I see. I assumed you had some experience and you need some for what I suggested.
In that case I think the best advice is just to make sure there is good overlap. The requirement is said to be "40 %", but I don't know how that number is arrived at (area or structure).
In my experience auto-align/photomerge works very well, both in CS4 and CS5, as long as the overlap is sufficient so that it has something to work with.
I decided to try just throwing them all in a new doc and aligning them by reducing opacity on one, moving the next over, return the opacity to 100%, etc. Considering they're all taken from a scope it works pretty well but it would have been nice to get the auo features to do the work.