Title :
Automatic selection of the region of interest in ancient scores
Author :
Barbancho, Isabel ; Segura, Cristian ; Tardón, Lorenzo J. ; Barbancho, Ana M.
Author_Institution :
Dept. de Ing. de Comun., Univ. de Malaga, Málaga, Spain
Abstract :
In this paper, an automatic system for the extraction of the region of interest in ancient scores is presented. The system is able to separate the image into two pages, when needed, then, using images that contain one page of score, the method performs a correction of the rotation before estimating the region of interest. For the estimation of the region of interest, three different methods have been developed: eroded edges method, wavelet method and eroded image and edge method. The three methods obtain quite good performance, being the wavelet based method the one that achieves best performance regarding execution time and accuracy in the process of selection of the region of interest.
Keywords :
feature extraction; history; image recognition; music; wavelet transforms; OMR; ancient scores; automatic system; eroded edges method; eroded image method; feature extraction; image separation; optical music recognition; region of interest; rotation correction; wavelet method; Books; Color; Data mining; Degradation; Digital images; Fungi; Ink; Lighting; Optical computing; Ordinary magnetoresistance;
Conference_Titel :
MELECON 2010 - 2010 15th IEEE Mediterranean Electrotechnical Conference
Conference_Location :
Valletta
Print_ISBN :
978-1-4244-5793-9
DOI :
10.1109/MELCON.2010.5476269