need help on final year project

[mcathymatthew]

hi everyone!!

I am a final year student doing software engineering..
I need help on my final year project. I am interested in music. i have an idea but i am not sure it can be finished in 7 months time or not. my plan is a sheet music reader. Asystem that is capble to scan a sheet of music and analyse it. XML file should be generated that captures as much as possible of the music information on the music sheet. Then generating MIDI output from the Music XML file. I really do not know how to do go about and do this project. How do i read the sheet music. Do i need to read the pixels to do that? What are the areas of study i need to consider? Please give me advice on this project.
I thank all of you in advance!!

[jwenting]

Start with doing extensive research into the subject, and then determine whether it's feasible for your experience and deadline (if you have to ask whether something like this is even possible I'd guess to conclude in advance that it's not feasible for you).

That sounds difficult. You would need to scan the sheet music using a scanner then store the picture in a bitmap or jpeg file.

The next part is the most difficult and requires some sort of image recognising software. For instance, how do you get the program to recognise the start of the bar, recognise crotchets semibreves, quavers and so on.
That leads you into the realms of AI, :eek: And that in itself will take some time to learn and even longer to devise sum kind of solution.

If u take out the AI part and assume the music is given in a discrete manner, I'd say sure, try it. That's what i think anyway. :-|

[tonakai]

after you scan your sheet, all you need to do is some sort of character recognation... you need to study some digital image processing book to learn character/pattern recognation.
simple recognation algorithms will work without problem if your sheets are not hand-written. if they are hand-written you need more complex algorithms...

Actually, I've had a quick look into this, and I think u might be able to do this if u use something like a piano roll as input--

http://en.wikipedia.org/wiki/Piano_roll

The principal works by reading perforations on the paper and depending on their alignment, will give the note pitch and speed.

Now the idea of using perforations would be a helluva lot easier to do than complete musical score recognition.

Think of it like getting your computer to convert braille into actual messages... Maybe, although u'll still need AI, kohen networks and all that malarky. :idea:

[hari_cseiitb]

hi everyone!!

I am a final year student doing software engineering..
I need help on my final year project. I am interested in music. i have an idea but i am not sure it can be finished in 7 months time or not. my plan is a sheet music reader. Asystem that is capble to scan a sheet of music and analyse it. XML file should be generated that captures as much as possible of the music information on the music sheet. Then generating MIDI output from the Music XML file. I really do not know how to do go about and do this project. How do i read the sheet music. Do i need to read the pixels to do that? What are the areas of study i need to consider? Please give me advice on this project.
I thank all of you in advance!!

Possible start point would be reading a .wmv file. and playing with the freqeuncies. Need to understand the signals and modifying them. Areas of Study : Digital Signal Processing.

one can store the wave features in a data structure and then modify and create the wave back.