An over-segmentation method for single-touching Chinese handwriting with learning-based filtering

Liang Xu*, Fei Yin, Qiu Feng Wang, Cheng Lin Liu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

18 Citations (Scopus)

Abstract

The segmentation of touching characters is still a challenging task, posing a bottleneck for offline Chinese handwriting recognition. In this paper, we propose an effective over-segmentation method with learning-based filtering using geometric features for single-touching Chinese handwriting. First, we detect candidate cuts by skeleton and contour analysis to guarantee a high recall rate of character separation. A filter is designed by supervised learning and used to prune implausible cuts to improve the precision. Since the segmentation rules and features are independent of the string length, the proposed method can deal with touching strings with more than two characters. The proposed method is evaluated on both the character segmentation task and the text line recognition task. The results on two large databases demonstrate the superiority of the proposed method in dealing with single-touching Chinese handwriting.

Original languageEnglish
Pages (from-to)91-104
Number of pages14
JournalInternational Journal on Document Analysis and Recognition
Volume17
Issue number1
DOIs
Publication statusPublished - Mar 2014
Externally publishedYes

Keywords

  • Chinese handwriting
  • Geometric features
  • Learning-based filtering
  • Over-segmentation
  • Single-touching strings

Fingerprint

Dive into the research topics of 'An over-segmentation method for single-touching Chinese handwriting with learning-based filtering'. Together they form a unique fingerprint.

Cite this