Optical Character Recognition for Music Notation

I am wondering if anyone out there has had any experience with Optical Character Recognition for Music Notation (or OCM).  It sounds like a really neat idea and am interested to try it out although it seems to me that there is a fairly large cost barrier to doing this and also that it takes a lot of time.  I just read Andrew Bullen’s article “Bringing Sheet Music to Life” My Experiences with OMR” in Code4Lib Journal – Issue 3.  Prior to reading this article I had not heard about this technology.  I think it would be neat project to digitize some of the sheet music in the collection in our archives at the library where I work.  Any thoughts, ideas or suggestions would be very much appreciated.