On Machine Understanding of Online Handwritten Mathematical Expressions

  • Authors:
  • Utpal Garain;B. B. Chaudhuri

  • Affiliations:
  • -;-

  • Venue:
  • ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper aims at automatic recognition of onlinehandwritten mathematical expressions written on anelectronic tablet. The proposed technique involves twomajor stages: symbol recognition and structural analysis.A multiple-classifier consists of both parametric and non-parametricclassifier has been used for recognition ofsymbols. Parametric classifier is based on HiddenMarkov Model (HMM), whereas, non-parametricclassifier uses Nearest Neighbor classification scheme.Structural analysis uses several online and offlinefeatures to identify the spatial relationships amongsymbols. A context free grammar has been designed toconvert input expressions into their corresponding Latexstrings. Contextual information has been used to correctseveral errors occurring at both recognition andstructural analysis stage. A new method for evaluatingperformance of the proposed systems has beenformulated. Experiments on a dataset of considerable sizeshow high efficiency of the proposed system.