Research and implementation of Android based optical music recognition

Summarization: As stated on Gerd Castan’s page about OMR, “Doing OMR is hard. Very hard. If you follow the OMR science links you will find much more theses about OMR than there are public available programs”. The motivation behind this thesis was to bring music to life. It is a great opportunity to have the software that can instantly produce a song from a sheet music in a course book. Presently, the respective commercial software is mainly available on a desktop platform, which requires a high-performance computer and a scanner to scan a sheet music. This project will survey the different techniques that have been used to perform OMR on printed music scores and an application by the name of AOMR will be developed. AOMR project renovates this kind of software by applying it on a mobile platform instead. This application, which highly involves image processing techniques and artificial intelligence applications, can recognize a scanned image of sheet music to be interpreted and exported as an audible representation via MIDI synthesis while handling resource utilization on an Android mobile phone platform. Limited processing performance and memory capacity, including the lack of image processing and other related APIs, are major issues that cause the algorithms used in the application to be different from traditional approaches applied in software on a PC platform. Such program is meant not only to be a simple OMR application but also the starting point for a future work about specialists having an interactive software to rely on for both as a practical tool for work and educational purposes