First of all, please see the open-source AForge.NET:
http://en.wikipedia.org/wiki/AForge.net[
^],
http://www.aforgenet.com/[
^].
Alternatively, consider Emgu CV:
http://www.emgu.com/wiki/index.php/Main_Page[
^].
This is a is a cross platform .Net wrapper to the OpenCV image processing library; and OpenCV is an open-source C++ computer vision library:
http://en.wikipedia.org/wiki/OpenCV[
^],
http://opencv.org/[
^].
[EDIT]
There are also some works closer to ORM. More exactly, it's OCR, but you can use it for OMR purposes (
http://en.wikipedia.org/wiki/Optical_mark_recognition[
^]).
Please see:
http://www.atalasoft.com/products/dotimage/ocr[
^],
http://www.pixel-technology.com/freeware/tessnet2/[
^].
Unfortunately, I must warn you: I never found any Open-Source OCR solutions which would be good enough. There are few CodeProject solutions I like, but, to best of knowledge, they are all incomplete, would need serious amount of work to make a working application out of them.
Please see the CodeProject articles:
Neural Network OCR[
^],
Creating Optical Character Recognition (OCR) applications using Neural Networks[
^],
Neural Network Classifier[
^],
A C# Project in Optical Character Recognition (OCR) Using Chain Code[
^],
Unicode Optical Character Recognition[
^],
Hebb neural network[
^],
Backpropagation Artificial Neural Network in C++[
^],
OCR Line Detection[
^].
I have to warn you: some of these works are very good, much better compared to the quality I complained about above, but none of these works is completed into a ready-to-use application. They all require considerable amount of work to complete to such level.
Maybe there are newer projects I'm not familiar with; so it would not hurt if you do your own search in CodeProject and elsewhere. Please, if you find something good, please reply in the comment to this answer.
Good luck,
—SA