For example, I have two objects, each with an image. The objects have an x and y, but they have more than just a width and height - they have a shape. I'm not talking about rectangular collision, but actual (reasonably) precise measurements that don't take a huge amount of load or run-time processing.
The best possible solution I could come up with would be to draw a blank object (off-screen, mind you), draw both objects, count up the number of background pixels remaining, and compare that with the total number of background pixels minus the sum of the number of pixels in the two objects.
That should give you an idea of what I am looking for. This method would be faster than some other methods I could think of, but I'm sure there must be a better way. I have seen many programs with something like this, and I would like to know how it could be done in a practical way.

