Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy

Toru Wakahara, Seiichi Uchida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper addresses the problem of how to extract, describe, and evaluate handwriting deformation from the deterministic viewpoint for improving recognition accuracy. The key ideas are threefold. The first is to extract handwriting deformation vector field (DVF) between a pair of input and target images by 2D warping. The second is to hierarchically decompose the DVF by a parametric deformation model of global/local affine transformation, where local affine transformation is iteratively applied to the DVF by decreasing window sizes. The third is to accept only low-order deformation components as natural, within-class handwriting deformation. Experiments using the handwritten numeral database IPTP CDROM1B show that correlation-based matching absorbing components of global affine transformation and local affine transformation up to the 3rd order achieved a higher recognition rate of 92.1% than that of 87.0% obtained by original 2D warping.

Original languageEnglish
Title of host publicationProceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010
Pages1860-1863
Number of pages4
DOIs
Publication statusPublished - Nov 18 2010
Event2010 20th International Conference on Pattern Recognition, ICPR 2010 - Istanbul, Turkey
Duration: Aug 23 2010Aug 26 2010

Publication series

NameProceedings - International Conference on Pattern Recognition
ISSN (Print)1051-4651

Other

Other2010 20th International Conference on Pattern Recognition, ICPR 2010
CountryTurkey
CityIstanbul
Period8/23/108/26/10

Fingerprint

Decomposition
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Wakahara, T., & Uchida, S. (2010). Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010 (pp. 1860-1863). [5597196] (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2010.459

Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy. / Wakahara, Toru; Uchida, Seiichi.

Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 1860-1863 5597196 (Proceedings - International Conference on Pattern Recognition).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wakahara, T & Uchida, S 2010, Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy. in Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010., 5597196, Proceedings - International Conference on Pattern Recognition, pp. 1860-1863, 2010 20th International Conference on Pattern Recognition, ICPR 2010, Istanbul, Turkey, 8/23/10. https://doi.org/10.1109/ICPR.2010.459
Wakahara T, Uchida S. Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 1860-1863. 5597196. (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2010.459
Wakahara, Toru ; Uchida, Seiichi. / Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy. Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. pp. 1860-1863 (Proceedings - International Conference on Pattern Recognition).
@inproceedings{792d22cde364463fb2e8920330555c4e,
title = "Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy",
abstract = "This paper addresses the problem of how to extract, describe, and evaluate handwriting deformation from the deterministic viewpoint for improving recognition accuracy. The key ideas are threefold. The first is to extract handwriting deformation vector field (DVF) between a pair of input and target images by 2D warping. The second is to hierarchically decompose the DVF by a parametric deformation model of global/local affine transformation, where local affine transformation is iteratively applied to the DVF by decreasing window sizes. The third is to accept only low-order deformation components as natural, within-class handwriting deformation. Experiments using the handwritten numeral database IPTP CDROM1B show that correlation-based matching absorbing components of global affine transformation and local affine transformation up to the 3rd order achieved a higher recognition rate of 92.1{\%} than that of 87.0{\%} obtained by original 2D warping.",
author = "Toru Wakahara and Seiichi Uchida",
year = "2010",
month = "11",
day = "18",
doi = "10.1109/ICPR.2010.459",
language = "English",
isbn = "9780769541099",
series = "Proceedings - International Conference on Pattern Recognition",
pages = "1860--1863",
booktitle = "Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010",

}

TY - GEN

T1 - Hierarchical decomposition of handwriting deformation vector field for improving recognition accuracy

AU - Wakahara, Toru

AU - Uchida, Seiichi

PY - 2010/11/18

Y1 - 2010/11/18

N2 - This paper addresses the problem of how to extract, describe, and evaluate handwriting deformation from the deterministic viewpoint for improving recognition accuracy. The key ideas are threefold. The first is to extract handwriting deformation vector field (DVF) between a pair of input and target images by 2D warping. The second is to hierarchically decompose the DVF by a parametric deformation model of global/local affine transformation, where local affine transformation is iteratively applied to the DVF by decreasing window sizes. The third is to accept only low-order deformation components as natural, within-class handwriting deformation. Experiments using the handwritten numeral database IPTP CDROM1B show that correlation-based matching absorbing components of global affine transformation and local affine transformation up to the 3rd order achieved a higher recognition rate of 92.1% than that of 87.0% obtained by original 2D warping.

AB - This paper addresses the problem of how to extract, describe, and evaluate handwriting deformation from the deterministic viewpoint for improving recognition accuracy. The key ideas are threefold. The first is to extract handwriting deformation vector field (DVF) between a pair of input and target images by 2D warping. The second is to hierarchically decompose the DVF by a parametric deformation model of global/local affine transformation, where local affine transformation is iteratively applied to the DVF by decreasing window sizes. The third is to accept only low-order deformation components as natural, within-class handwriting deformation. Experiments using the handwritten numeral database IPTP CDROM1B show that correlation-based matching absorbing components of global affine transformation and local affine transformation up to the 3rd order achieved a higher recognition rate of 92.1% than that of 87.0% obtained by original 2D warping.

UR - http://www.scopus.com/inward/record.url?scp=78149489513&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149489513&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2010.459

DO - 10.1109/ICPR.2010.459

M3 - Conference contribution

AN - SCOPUS:78149489513

SN - 9780769541099

T3 - Proceedings - International Conference on Pattern Recognition

SP - 1860

EP - 1863

BT - Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010

ER -