Title :
Detecting Main Body Size in Document Images
Author :
Diamantatos, Paraskevas ; Verras, Vasileios ; Kavallieratou, Ergina
Author_Institution :
Dept. Inf. & Commun. Syst. Eng., Univ. of the Aegean, Samos, Greece
Abstract :
In this paper, two techniques are presented, appropriate to detect the text main body size in a document image. One measures it directly, while the other estimates the baselines first. Both are segmentation free. Experimental results are presented over a collection of handwritten text, as well as for a small collection of 10 printed document images, in order to give more objective results.
Keywords :
document image processing; image segmentation; document image segmentation; handwritten text; printed document images; text main body size; Estimation; Handwriting recognition; Hidden Markov models; Histograms; Image segmentation; Text analysis; baseline detection; document image processing; historical document images; word main body estimation;
Conference_Titel :
Document Analysis and Recognition (ICDAR), 2013 12th International Conference on
Conference_Location :
Washington, DC
DOI :
10.1109/ICDAR.2013.235
