An Automatic Mathematical Expression Understanding System

  • Authors:

  • Y. Guo;L. Huang;C. Liu;X. Jiang

  • Affiliations:

  • Chinese Academy of Sciences, Beijing 100080, China;Chinese Academy of Sciences, Beijing 100080, China;Chinese Academy of Sciences, Beijing 100080, China;Chinese Academy of Sciences, Beijing 100080, China

  • Venue:
  • ICDAR '07 Proceedings of the Ninth International Conference on Document Analysis and Recognition - Volume 02
  • Year:
  • 2007

  • Mathematical Symbol Indexing

    AI*IA '09: Proceedings of the XIth International Conference of the Italian Association for Artificial Intelligence Reggio Emilia on Emergent Perspectives in Artificial Intelligence

Quantified Score

Hi-index 0.00

Visualization

Abstract

An automatic mathematical expression (ME) understanding system is presented. This system is consisted of three procedures, i.e., pre-processing, symbol segmentation, structure analysis among symbols within the ME. In symbol segmentation procedure, a three-stage segmentation method based on dynamic programming (DP) is introduced to segment symbols in the ME image. In structure analysis procedure, a ME is decomposed into several basic expressions based on reconstructing the ME global structure. Experimental results on about 3268 ME images showed that the presented system worked well. Top-1 ME recognition accuracy reached 78.2%.