NN Music Score Recognition

Head and Stem Extraction from Printed Music Scores

H. Miyao and Y. Nakano
Proceedings of 3rd ICDAR, pp.1074-1079 (1995)

Abstract

In an automatic music score recognition system, it is very important to extract heads and stems of notes, since these symbols are most ubiquitous in a score and musically important. The purpose of our system is to present an accurate and speedy extraction of note heads (except the whole notes and stems according to the following procedures.

(1) We extract all regions which are considered as candidates of stems or heads.
(2) To identify heads from the candidates, we use a three-layer neural network.
(3) The weights for the network are learned by the back propagation method.
In the learning the network learns the spatial constraints between the heads and surroundings rather than the shapes of heads.
(4) After the learning process us completed we use this network to identify a number of test head candidates.
(5) The stem candidates touching the detected heads are extracted as true stems.
As an experimental result, we obtained high recognition rates of 99.0% and 99.2% for stems and note heads, respectively. It took time from about 40 to 100 seconds to do for a printed piano score on A4 sheet using a workstation. Therefore, our system can analyze it 10 times or as faster as the manual work.

[Music Score Recognition] [Research Themes of Prof. Nakano.]

mail address: <- Please enter the string into the address field

First Written Before June 17, 1998
Transplanted to KSU Before June 18, 2003
Transplanted to So-net April 22, 2007
Last Update April 22, 2007