Publications

Abstract : Most of the popular music compositions exists in either hand-written or facsimile format even in this modern era. In order to preserve these precious music sheets, they needs to be digitization followed by conversion into machine readable format, such that a system can decode the symbolic images automatically and play the music. If an image of the musical work becomes available, the entire process of digitization can be considered as a document image processing application. Musical notes to it machine readable format conversion mainly aims at reading automatically generated musical scores in order to convert them to an electronic format such as MIDI file. In this paper, we propose a system that takes an image of a music sheet as input, identifies music symbols, calculates their pitch and duration and outputs a string corresponding to each line in the music sheet which can be played using Musicgui in DSP Toolbox. Major challenge faced is the removal of staff lines in the music sheet image without losing music symbols. Major steps involved in this are staff line removal, symbol extraction, symbol recognition, and conversion of musical scores into MIDI format.