Because usable sensor size will depend on how much of the image is overlayed in the two conditions. Could add arrows indicating which way the images need to be translated, for example. Or simply report overlap between reference and target. That would also work nicely.