Well the first thing I'd suggest is forget about the mouse: it doesn't sounds like you need to manipulate the actual mouse position with a video feed.  Instead replace the references to mouse position in your code with x and y variables (you can always pass the mouse coordinates to these variables to test it ).

Then get your head round JMyron.  I'd suggest doing this in a separate sketch: that way you can concentrate on figuring it out without having to fit it into your existing code.  Use the reference.  Also make sure you search the forum and exhibitions for uses of JMyron

It looks to me like you'd be interested in differenceImage() and globCenters()...  the only problem being that it's likely you'll be dealing with more than one glob at any given time - so you'll also need to work out how to choose which one to follow...  but once you've done that it should just be a matter of passing the coordinates to your x/y variables in your code