Content features for logical document labeling
Jian Liang, David Doermann · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2003
The use of content feature extracted from recognized text is valuable in labeling logical elements in documents without rigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experiment results are presented.