Title :
Features for printed document image analysis
Author :
Duong, Jean ; Emptoz, Hubert ; Côté, Myriam
Author_Institution :
Lab. de Reconnaissance de Formes et Vision, Inst. Nat. des Sci. Appliquees de Lyon, Villeurbanne, France
Abstract :
This paper presents features for text/non-text area separation in printed document images. First, it introduces entropic discrimination, i.e., a simple separation using only one feature. Then, a brief recall on existing texture and geometric discriminant parameters proposed in previous research (2001, 2002) is included. Several of them are statistically examined.
Keywords :
document image processing; entropy; feature extraction; learning (artificial intelligence); pattern classification; document zone classification; entropic discrimination; entropy; feature extraction; geometric discriminant parameters; printed document image analysis; text separation; training set; Entropy; Graphics; Histograms; Image analysis; Image texture analysis; Information analysis; Labeling; Optical character recognition software; Reconnaissance; Text analysis;
Conference_Titel :
Pattern Recognition, 2002. Proceedings. 16th International Conference on
Print_ISBN :
0-7695-1695-X
DOI :
10.1109/ICPR.2002.1047840
