Besides the manual analysis of handwritten music scores, another approach for automatic handwriting analysis has been considered by the project partner Fraunhofer Institute. The goal is to be able to extract the features of handwritten music scores by means of computer-aided image processing functions. This process consists of several steps.
During image processing, techniques are applied in order to reduce image noise and separate background from foreground and other procedures were done to ensure the following analysis. Afterwards, the staff, the primitives like note heads and stems, bars and other complex symbols will be detected. An exact object detection occurs by comparing all primitives and appropriately assembling them. Afterwards, an analysis of the handwriting characteristics can take place using detailed measurements.
In order to support automatic handwriting analysis, a software system is currently being designed and developed (see right graphic) that will grow to become a complex analysis tool during the course of the project. Please take a look at the reports of the Fraunhofer Institute in Rostock for further detailed information about special analysis techniques for handwriting characteristics.