I'm not sure if you're referring to the tansit technique where a planet passes in front of a star multiple times which allows to determine the size and distance

I'm not sure if you're referring to the tansit technique where a planet passes in front of a star multiple times which allows to determine the size and distance from the detectable dips in brightness. This technique wouldn't work well with a backdrop since it never passes in front of the same star twice. I'm also not sure if we can resolve the planet optically i.e. 1) it will always be smaller than a pixel of the telescope sensor 2) the pixels of the telescope sensor always collect light from the backdrop even if the planet is dead center on the group of pixels. When you detect the light of a planet without backdrop both problems would be much smaller because then you can actually see some tiny dark blurry object moving relative to the stars around it in the image.