musical notes reader Anna Shmushkin & Lior Abramov

Introduction Optical music recognition (OMR) has been the subject of research for decade. Many image processing algorithms and techniques have been developed to address this problem, and yet the problem still poses many challenges to scientists and researchers today. The goal of this project is to supply playback mechanism for the parsed musical notes.

UI

image rectification Approach and Method

Staff Lines  Detection:  Detection: The first step in processing a given input image is to detect the individual staff lines of the piece of music.  Parameter Extraction:  Parameter Extraction: Once we had the final staff line locations from the previous step, we calculated the gap between the staff lines, g, by once again computing the median of the set of interval lengths between adjacent staff lines.  Removal:  Removal: We have scanned across the rows of the staff lines, removing the existing black pixel if there are no black pixels above nor below it.

Segmentation  We have preformed x-transformation in order to segmented the image to connected group segmentation and note segmentation. Note Head Detection  We were able to identify the coordinates of the note heads using the best note template matching we could find. (Using Matlab normalized 2-D cross-correlation function).

Note Identification  Classified the note type, octave, and pitch of the note. To determine the octave and pitch, we used the centroid of the region found previously and crossreferenced it with the staff line locations of the original image in order to round the position to the nearest half step of g, which directly was converted into an octave and pitch.

Notes to music  Notes to Lilypond input  Lilypond does the rest Lilypond input

Results and Conclusions This has been our first real experience to try and grasp concepts out of images on our own accord. The goal we set to ourselves at the beginning of this journey was high yet feasible. we attacked the issue little by little, making up heuristics and asserting them to real life as we went along. There is still much work to be done and a whole set of improvements yet to be made, but as a first stage, we think we have managed to make some nice steps along our own Mount Improbable, or rather, probable.