Title :
A rule learning method for academic document image processing
Author :
Takasu, Atsuhiro ; Satoh, Shinichi ; Katsura, Eishi
Author_Institution :
Res. & Dev. Dept., Nat. Center for Sci. Inf. Syst., Tokyo, Japan
Abstract :
A syntactic rule learning method is presented for analyzing document images and constructing a database from them. This method is used in a digital library system named CyberMagazine, where document images are sequentially converted into database tuples by block segmentation, rough classification, and syntactic analysis. The syntactic rule has an ability to analyze symbols located in two dimensional plane, and has a syntax similar to an ordinal context free grammar except for the concatenation of symbols. In the presented learning method, the syntactic rules are generated from a set of parse trees by decomposing the trees according to non terminal symbols, generalizing the decomposed trees to a syntactic rule, and merging them
Keywords :
computational linguistics; context-free grammars; document image processing; educational administrative data processing; image classification; image segmentation; learning (artificial intelligence); trees (mathematics); visual databases; CyberMagazine; academic document image processing; block segmentation; database tuples; decomposed trees; digital library system; non terminal symbols; ordinal context free grammar; parse trees; rough classification; symbol concatenation; syntactic analysis; syntactic rule learning method; two dimensional plane; visual database; Classification tree analysis; Decision trees; Document image processing; Image analysis; Image converters; Image databases; Image processing; Image segmentation; Learning systems; Research and development;
Conference_Titel :
Document Analysis and Recognition, 1995., Proceedings of the Third International Conference on
Conference_Location :
Montreal, Que.
Print_ISBN :
0-8186-7128-9
DOI :
10.1109/ICDAR.1995.598985
