Deep Neural Networks for Document Processing of Music Score Images

There is an increasing interest in the automatic digitization of medieval music documents. Despite efforts in this field, the detection of the different layers of information on these documents still poses difficulties. The use of Deep Neural Networks techniques has reported outstanding results in m...

Full description

Bibliographic Details
Main Authors:	Jorge Calvo-Zaragoza, Francisco J. Castellanos, Gabriel Vigliensoni, Ichiro Fujinaga
Format:	Article
Language:	English
Published:	MDPI AG 2018-04-01
Series:	Applied Sciences
Subjects:	Optical Music Recognition music document processing music score images Medieval manuscripts convolutional neural networks
Online Access:	http://www.mdpi.com/2076-3417/8/5/654

id	doaj-a3735d4d4a9c4092a390eb758d182006
record_format	Article
spelling	doaj-a3735d4d4a9c4092a390eb758d1820062020-11-24T21:14:24ZengMDPI AGApplied Sciences2076-34172018-04-018565410.3390/app8050654app8050654Deep Neural Networks for Document Processing of Music Score ImagesJorge Calvo-Zaragoza0Francisco J. Castellanos1Gabriel Vigliensoni2Ichiro Fujinaga3PRHLT Research Center, Universitat Politècnica de València, 46022 Valencia, SpainSoftware and Computing Systems, University of Alicante, 03690 Alicante, SpainSchulich School of Music, McGill University, Montreal, QC H3A 0G4, CanadaSchulich School of Music, McGill University, Montreal, QC H3A 0G4, CanadaThere is an increasing interest in the automatic digitization of medieval music documents. Despite efforts in this field, the detection of the different layers of information on these documents still poses difficulties. The use of Deep Neural Networks techniques has reported outstanding results in many areas related to computer vision. Consequently, in this paper, we study the so-called Convolutional Neural Networks (CNN) for performing the automatic document processing of music score images. This process is focused on layering the image into its constituent parts (namely, background, staff lines, music notes, and text) by training a classifier with examples of these parts. A comprehensive experimentation in terms of the configuration of the networks was carried out, which illustrates interesting results as regards to both the efficiency and effectiveness of these models. In addition, a cross-manuscript adaptation experiment was presented in which the networks are evaluated on a different manuscript from the one they were trained. The results suggest that the CNN is capable of adapting its knowledge, and so starting from a pre-trained CNN reduces (or eliminates) the need for new labeled data.http://www.mdpi.com/2076-3417/8/5/654Optical Music Recognitionmusic document processingmusic score imagesMedieval manuscriptsconvolutional neural networks
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Jorge Calvo-Zaragoza Francisco J. Castellanos Gabriel Vigliensoni Ichiro Fujinaga
spellingShingle	Jorge Calvo-Zaragoza Francisco J. Castellanos Gabriel Vigliensoni Ichiro Fujinaga Deep Neural Networks for Document Processing of Music Score Images Applied Sciences Optical Music Recognition music document processing music score images Medieval manuscripts convolutional neural networks
author_facet	Jorge Calvo-Zaragoza Francisco J. Castellanos Gabriel Vigliensoni Ichiro Fujinaga
author_sort	Jorge Calvo-Zaragoza
title	Deep Neural Networks for Document Processing of Music Score Images
title_short	Deep Neural Networks for Document Processing of Music Score Images
title_full	Deep Neural Networks for Document Processing of Music Score Images
title_fullStr	Deep Neural Networks for Document Processing of Music Score Images
title_full_unstemmed	Deep Neural Networks for Document Processing of Music Score Images
title_sort	deep neural networks for document processing of music score images
publisher	MDPI AG
series	Applied Sciences
issn	2076-3417
publishDate	2018-04-01
description	There is an increasing interest in the automatic digitization of medieval music documents. Despite efforts in this field, the detection of the different layers of information on these documents still poses difficulties. The use of Deep Neural Networks techniques has reported outstanding results in many areas related to computer vision. Consequently, in this paper, we study the so-called Convolutional Neural Networks (CNN) for performing the automatic document processing of music score images. This process is focused on layering the image into its constituent parts (namely, background, staff lines, music notes, and text) by training a classifier with examples of these parts. A comprehensive experimentation in terms of the configuration of the networks was carried out, which illustrates interesting results as regards to both the efficiency and effectiveness of these models. In addition, a cross-manuscript adaptation experiment was presented in which the networks are evaluated on a different manuscript from the one they were trained. The results suggest that the CNN is capable of adapting its knowledge, and so starting from a pre-trained CNN reduces (or eliminates) the need for new labeled data.
topic	Optical Music Recognition music document processing music score images Medieval manuscripts convolutional neural networks
url	http://www.mdpi.com/2076-3417/8/5/654
work_keys_str_mv	AT jorgecalvozaragoza deepneuralnetworksfordocumentprocessingofmusicscoreimages AT franciscojcastellanos deepneuralnetworksfordocumentprocessingofmusicscoreimages AT gabrielvigliensoni deepneuralnetworksfordocumentprocessingofmusicscoreimages AT ichirofujinaga deepneuralnetworksfordocumentprocessingofmusicscoreimages
_version_	1716747382484369408

Deep Neural Networks for Document Processing of Music Score Images

Similar Items