Diacritical processing for unconstrained online handwriting recognition using a forward search

Out-of-order diacriticals introduce significant complexity to the design of an online handwriting recognizer, because they require some reordering of the time domain information. It is common in cursive writing to write the body of an `i' or `t' during the writing of the word, and then to return and dot or cross the letter once the word is complete. The difficulty arises because we have to look ahead, when scoring one of these letters, to find the mark occurring later in the writing stream that completes the letter. We should also remember that we have used this mark, so that we don't use it again for a different letter, and we should also penalize a word if there are some marks that look like diacriticals that are not used. One approach to this problem is to scan the writing some distance into the future to identify candidate diacriticals, remove them in a preprocessing step, and associate them with the matching letters earlier in the word. If done as a preliminary operation, this approach is error-prone: marks that are not diacriticals may be incorrectly identified and removed, and true diacriticals may be skipped. This paper describes a novel extension to a forward search algorithm that provides a natural mechanism for considering alternative treatments of potential diacriticals, to see whether it is better to treat a given mark as a diacritical or not, and directly compare the two outcomes by score. Received October 30, 1998 / Revised January 25, 1999     

Large vocabulary recognition of on-line handwritten cursive words

This paper presents a writer independent system for large vocabulary recognition of on-line handwritten cursive words. The system first uses a filtering module, based on simple letter features, to quickly reduce a large reference dictionary (lexicon) to a more manageable size; the reduced lexicon is subsequently fed to a recognition module. The recognition module uses a temporal representation of the input, instead of a static two-dimensional image, thereby preserving the sequential nature of the data and enabling the use of a Time-Delay Neural Network (TDNN); such networks have been previously successful in the continuous speech recognition domain. Explicit segmentation of the input words into characters is avoided by sequentially presenting the input word representation to the neural network-based recognizer. The outputs of the recognition module are collected and converted into a string of characters that is matched against the reduced lexicon using an extended Damerau-Levenshtein function. Trained on 2,443 unconstrained word images (11 k characters) from 55 writers and using a 21 k lexicon we reached a 97.9% and 82.4% top-5 word recognition rate on a writer-dependent and writer-independent test, respectively     

Severe meningoencephalitis in Borrelia burgdorferi infection]

A gas chromatographic-mass spectrometric method was used to separate quinine and its metabolites present in urine after oral dosing of 300 mg quinine in humans. The technique allowed the separation of quinine and ten metabolites. Four of these metabolites were definitely identified as 3-hydroxyquinine, 2'-quinone, O-desmethylquinine and 10,11-dihydroxydihydroquinine, by comparing their methane chemical ionization mass spectra with those of authentic standards prepared by organic synthesis. Six other metabolites are described for the first time in human urine. From their electron impact and chemical ionization mass spectra, we propose these compounds to be 3-hydroxy-2'-quinone, O-desmethyl-2'-quinone, O-desmethyl-3-hydroxyquinine, O-desmethyl-3-hydroxy-2'-quinone, 10,11-dihydroxydihydro-2'-quinone and 10,11-dihydroxydihydro-O-desmethylquinine. These secondary metabolites probably arose from further biotransformation of the four primary metabolites.     

Coding efficiency of multi-ring and single-ring differential chaincoding for telewriting application

The availability of small personal digital assistants (PDAs) with a touchscreen and communication capabilities has been an influential factor in the renewed interest in telewriting, a technique for the exchange of handwritten information through telecommunications means. In this context, differential chain coding algorithms for compression of the handwritten ink are revisited. In particular, it is shown that the coding efficiency of multi-ring differential chain coding (MRDCC) is not always better when compared to single ring differential chain coding (DCC), as previously suggested. These algorithms were tested on over 300 handwritten messages using a relative compactness criterion and a per-length distortion measure. The probabilities of relative vectors in MRDCC and DCC are related, an expression for relative compactness in the MRDCC case is introduced, and the application of Freeman's (1974) criteria for the selection of the appropriate code for a family of curves is illustrated