Resumen de Writer Identification by a Combination of Graphical Features in the Framework of Old Handwritten Music Scores
Alicia Fornés Bisquerra 
The analysis and recognition of historical document images has attracted growing interest in the last years. Mass digitization and document image understanding allows the preservation, access and indexation of this artistic, cultural and technical heritage. The analysis of handwritten documents is an outstanding subfield. The main interest is not only the transcription of the document to a standard format, but also, the identification of the author of a document from a set of writers (namely writer identification). Writer identification in handwritten text documents is an active area of study, and the literature is prolific in noteworthy contributions. However, the identification of the writer of graphical documents is still a challenge. The main objective of this thesis is the identification of the writer in old music scores, as an example of graphic documents. Concerning old music scores, many historical archives contain a huge number of sheets of musical compositions without information about the composer, and the research on this field could be helpful for musicologists. The writer identification framework proposed in this thesis combines three different writer identification approaches, which are the main scientific contributions. The first one is based on symbol recognition methods. For this purpose, two novel symbol recognition methods are proposed for coping with the typical distortions in hand-drawn symbols. The first one is a Dynamic Time Warping (DTW) based method, in which symbols are described by vector sequences, and a variation of the DTW-distance is used for computing the matching distance. The second one is called the Blurred Shape Model (BSM), in which a symbol is described by a probability density function that encodes the probability of pixel densities of image regions. The second writer identification approach preprocesses the music score for obtaining music lines, and extracts information about the slant, width of the writing, connected components, contours and fractals. Then, a k-NN classifier is the used to categorize the document image. Finally, the third approach extracts global information about the writing, by generating texture images from the music scores and extracting textural features. The feature space is defined in terms of Gabor filters and co-ocurence matrices. The high identification rates obtained in the experimental results demonstrate the suitability of the proposed ensemble architecture for the identification of the writer in music scores. To the best of our knowledge, this work is the first contribution on writer identification from images containing graphical languages.
