A perturbation model for generating synthetic text lines from existing cursively handwritten lines of text produced by human writers is presented. Our purpose is to improve the performance of an HMM-based off-line cursive handwriting recognition system by providing it with additional synthetic training data. Two kinds of perturbations are applied, geometrical transformations and thinning/thickening operations. The proposed perturbation model is evaluated under different experimental conditions.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Generation of synthetic training data for an HMM-based handwriting recognition system


    Contributors:
    Varga, T. (author) / Bunke, H. (author)


    Publication date :

    2003-01-01


    Size :

    449723 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Generation of Synthetic Training Data for an HMM-Based Handwriting Recognition System

    Varga, T. / Bunke, H. / Institute of Electrical and Electronics Engineers | British Library Conference Proceedings | 2003


    ABJAD HAWWAZ: An Offline Arabic Handwriting Recognition System

    Haraty, R. A. / El-Zabadani, H. M. | British Library Online Contents | 2005


    Rejection measures for handwriting sentence recognition

    Marukatat, S. / Artieres, T. / Gallinari, P. et al. | IEEE | 2002


    Principles of Handwriting Recognition in FineReader

    Tereshchenko, V. / Rybkin, V. / Shamis, A. et al. | British Library Online Contents | 1998


    On recognition of handwriting Chinese characters

    Li, C. C. / Sze, T. W. / Teng, T. L. et al. | NTRS | 1967