Advances in document image analysis : First Brazilian Symposium, BSDIAʾ97, Curitiba, Brazil, November 2-5, 1997 : proceedings

Handwriting recognition: State of the art and future trends.- Crucial combinations for the recognition of handwritten letters.- Recognition of printed and handwritten Arabic characters.- Objective evaluation of the discriminant power of features in an HMM-based word recognition system.- Effect of variability on letters generation with the vectorial delta-lognormal model.- Off-line signature verification: Recent advances and perspectives.- Research advances in graphics recognition: An update.- Future trends in retrospective document conversion.- Using binary pyramids to create multiresolution shape descriptors.- Contour pixel classification for character skeletonization.- Constraint propagation vs syntactical analysis for the logical structure recognition of library references.- Grammatical formalism for document understanding system: From document towards HTML text.- Document modeling for form class identification.- Composite document analysis by means of typographic characteristics.- Techniques for the automated testing of document analysis algorithms.- Graphical tools and techniques for querying document image databases.- Handwritten numeral recognition via fuzzy logic and local discriminating features.- Unsupervised learning of character prototypes.- Off-line cursive word recognition with a hybrid neural-HMM system.- A fuzzy perception for off-line handwritten signature verification.- An evaluation of different handwriting observation techniques from a signature verification point of view.- Generation of signatures by deformations.- An application of the sequential dynamic programming matching method to off-line signature verification.- A system for automatic form reading.- Automatic extraction of filled-in information from bankchecks based on prior knowledge about layout structure.- A simple methodology to bank cheque segmentation.