본문 바로가기
  • Home

Effective Handwriting Verification through DTW and PCA

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2009, 14(7), pp.25-32
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science

장석우 1 허문행 1 Gye-young Kim 2

1안양대학교
2숭실대학교

Accredited

ABSTRACT

In this paper, we propose a new handwriting verification method using pattern analysis in off-line environments. The proposed method first segments character regions in a document and extracts effective features from the segmented regions. It then estimates the similarity between the extracted non-linear features and reference ones by using dynamic time warping and principal component analysis. Our handwriting verification method extracts handwriting features effectively and enables the verification of handwriting with various lengths of features as well as ones of short patterns. The experimental results show that our method outperforms others in terms as accuracy. We expect that the proposed method will automate the manual handwriting verification tasks and provide much objectivity on handwriting identification.

Citation status

* References for papers published after 2022 are currently being built.

This paper was written with support from the National Research Foundation of Korea.