Home > Research > Publications & Outputs > Improving optical music recognition by combinin...

Electronic data

  • ImprovingOMR

    Final published version, 573 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

View graph of relations

Improving optical music recognition by combining outputs from multiple sources

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Improving optical music recognition by combining outputs from multiple sources. / Padilla Martin-Caro, Victor Manuel; McLean, Alex; Marsden, Alan Alexander et al.
Proceedings of the 16th International Society for Music Information Retrieval Conference. ed. / Meinard Mueller; Frans Wiering. 2015. p. 517-523.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Padilla Martin-Caro, VM, McLean, A, Marsden, AA & Ng, K 2015, Improving optical music recognition by combining outputs from multiple sources. in M Mueller & F Wiering (eds), Proceedings of the 16th International Society for Music Information Retrieval Conference. pp. 517-523. <http://ismir2015.uma.es/articles/187_Paper.pdf>

APA

Padilla Martin-Caro, V. M., McLean, A., Marsden, A. A., & Ng, K. (2015). Improving optical music recognition by combining outputs from multiple sources. In M. Mueller, & F. Wiering (Eds.), Proceedings of the 16th International Society for Music Information Retrieval Conference (pp. 517-523) http://ismir2015.uma.es/articles/187_Paper.pdf

Vancouver

Padilla Martin-Caro VM, McLean A, Marsden AA, Ng K. Improving optical music recognition by combining outputs from multiple sources. In Mueller M, Wiering F, editors, Proceedings of the 16th International Society for Music Information Retrieval Conference. 2015. p. 517-523

Author

Padilla Martin-Caro, Victor Manuel ; McLean, Alex ; Marsden, Alan Alexander et al. / Improving optical music recognition by combining outputs from multiple sources. Proceedings of the 16th International Society for Music Information Retrieval Conference. editor / Meinard Mueller ; Frans Wiering. 2015. pp. 517-523

Bibtex

@inproceedings{4d6325d8d2744106a288e9832b7d55b2,
title = "Improving optical music recognition by combining outputs from multiple sources",
abstract = "Current software for Optical Music Recognition (OMR) produces outputs with too many errors that render it an unrealistic option for the production of a large corpus of symbolic music files. In this paper, we propose a system which applies image pre-processing techniques to scans of scores and combines the outputs of different commercial OMR programs when applied to images of different scores of the same piece of music. As a result of this procedure, the combined output has around 50% fewer errors when compared to the output of any one OMR program. Image pre-processing splits scores into separate movements and sections and removes ossia staves which confuse OMR software. Post-processing aligns the outputs from different OMR programs and from different sources, rejecting outputs with the most errors and using majority voting to determine the likely correct details. Our software produces output in MusicXML, concentrating on accurate pitch and rhythm and ignoring grace notes. Results of tests on the six string quartets by Mozart dedicated to Joseph Haydn and the first six piano sonatas by Mozart are presented, showing an average recognition rate of around 95%.",
author = "{Padilla Martin-Caro}, {Victor Manuel} and Alex McLean and Marsden, {Alan Alexander} and Kia Ng",
year = "2015",
month = oct,
day = "30",
language = "English",
pages = "517--523",
editor = "Meinard Mueller and Frans Wiering",
booktitle = "Proceedings of the 16th International Society for Music Information Retrieval Conference",

}

RIS

TY - GEN

T1 - Improving optical music recognition by combining outputs from multiple sources

AU - Padilla Martin-Caro, Victor Manuel

AU - McLean, Alex

AU - Marsden, Alan Alexander

AU - Ng, Kia

PY - 2015/10/30

Y1 - 2015/10/30

N2 - Current software for Optical Music Recognition (OMR) produces outputs with too many errors that render it an unrealistic option for the production of a large corpus of symbolic music files. In this paper, we propose a system which applies image pre-processing techniques to scans of scores and combines the outputs of different commercial OMR programs when applied to images of different scores of the same piece of music. As a result of this procedure, the combined output has around 50% fewer errors when compared to the output of any one OMR program. Image pre-processing splits scores into separate movements and sections and removes ossia staves which confuse OMR software. Post-processing aligns the outputs from different OMR programs and from different sources, rejecting outputs with the most errors and using majority voting to determine the likely correct details. Our software produces output in MusicXML, concentrating on accurate pitch and rhythm and ignoring grace notes. Results of tests on the six string quartets by Mozart dedicated to Joseph Haydn and the first six piano sonatas by Mozart are presented, showing an average recognition rate of around 95%.

AB - Current software for Optical Music Recognition (OMR) produces outputs with too many errors that render it an unrealistic option for the production of a large corpus of symbolic music files. In this paper, we propose a system which applies image pre-processing techniques to scans of scores and combines the outputs of different commercial OMR programs when applied to images of different scores of the same piece of music. As a result of this procedure, the combined output has around 50% fewer errors when compared to the output of any one OMR program. Image pre-processing splits scores into separate movements and sections and removes ossia staves which confuse OMR software. Post-processing aligns the outputs from different OMR programs and from different sources, rejecting outputs with the most errors and using majority voting to determine the likely correct details. Our software produces output in MusicXML, concentrating on accurate pitch and rhythm and ignoring grace notes. Results of tests on the six string quartets by Mozart dedicated to Joseph Haydn and the first six piano sonatas by Mozart are presented, showing an average recognition rate of around 95%.

M3 - Conference contribution/Paper

SP - 517

EP - 523

BT - Proceedings of the 16th International Society for Music Information Retrieval Conference

A2 - Mueller, Meinard

A2 - Wiering, Frans

ER -