eNoteHistory
Subproject Image Processing Homepage
Objectives
- To develop a software system that facilitates the work of the
music historians in identifying the scribe of handwritten music scores.
- Optical Music Recognition (OMR) is similar to Opical Character
Recognition (OCR) except that we try to detect the musical notation
symbols such as notes, clefs, pauses etc.
- OMR is still in its infancy. While some software systems exist
for OCR on printed music scores, the recognition of handwritten music
scores remains an ongoing research topic. We aim at developing
algorithm that robustly recognise the individual notation symbols,
irrespective of the particular handwriting style of a scribe.
Current Work
- Our approach is based on a hierarchical algorithm with four
levels.
- In the first level, the digitised scores undergo a
preprocessing which consists of smoothing, histogram equalisation,
segmentation, and morphological operations.
- In the second level, the preprocessed images are analysed by
splitting the segmented foreground information into five layers. The
layers contain primitives that may form music notation symbols.
- In the third level, the information from the layers is used
to recognise music notation symbols of interest. Information about the
objects is collected in a set of characteristics.
- Finally, a classification of the given score page is
carried out based on the information in the set.
- We have built a testbed using Java Advanced Imaging (JAI), in
which we test various methods for each of the four levels. Here's a
snapshot of the ENoteImageAnalyser:
ENoteImageAnalyser - A testbed for
image processing methods
Publications
- R. Göcke. Building a System
for Writer Identification on
Handwritten Music Scores. Proceedings of the IASTED
International Conference on Signal
Processing, Pattern Recognition, and Applications SPPRA 2003,
pp.250-255, Rhodes, Greece, 30 June – 2 July 2003 .pdf (125KB)
- R. Göcke, J. Voskamp, E. Krüger, T. Schwinger, I. Bruder, A.
Finger, A. Heuer, T. Ignatova. Ein System zur Identifikation der
Schreiber von historischen Musikhandschriften. 4. Information- und
Kommunikations-Tage Mecklenburg-Vorpommern, CD-ROM only, Rostock,
Germany, 18-20 June 2003 .pdf (84KB)
(in German)
Some Interesting Links
Project
page at the Fraunhofer IGD Rostock
Kia Ng's homepage (Optical Manuscript
Analysis)
Interdiscplinary Centre of
Scientific Research in Music (Uni Leeds, UK)
Wedelmusic - Web Delivering of
Music
Optical Music Recognition
Bibliography
Dorothea Blostein's
homepage (OCR, OMR)
Last update 24 Oct 2003