Handwritten Character Recognition Using Combined Gradient and Wavelet Feature

Weipeng Zhang; Yuan Yan Tang; Yun Xue

doi:10.1109/ICCIAS.2006.294218

Handwritten Character Recognition Using Combined Gradient and Wavelet Feature

Weipeng Zhang^*
, Yuan Yan Tang
, Yun Xue

^*Corresponding author for this work

Department of Mathematics

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

12 Citations (Scopus)

Abstract

A novel feature based on the combination of gradient feature and coefficients of wavelet transform is developed in this paper. In handwritten character recognition, the gradient feature represents local characteristic properly, but it is sensitive to the deformation of handwritten character. Meanwhile, wavelet transform represents the character image in multiresolution analysis and keeps adequate global characteristic in different scales. In order to improve the discrimination power, we composed local and global characteristic in a combined feature. Three combination schemes are described in this paper. Experiments are conducted on two Chinese character databases, ETL8B subset (197 categories) and HKBU-SC110 (110 categories), to test the performance of proposed feature. The recognition accuracies of our feature achieve 95.53% and 93.77% for ETL8B subset and HKBU-SC110 by 1-NN classifier, respectively, which are higher than those of gradient feature.

Original language	English
Title of host publication	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006
Editors	Yiu-ming Cheung, Yuping Wang, Hailin Liu
Publisher	IEEE
Pages	662-667
Number of pages	6
ISBN (Print)	1424406056, 1424406048, 9781424406050
DOIs	https://doi.org/10.1109/ICCIAS.2006.294218
Publication status	Published - Nov 2006
Event	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006 - Guangzhou, China Duration: 3 Nov 2006 → 6 Nov 2006 https://ieeexplore.ieee.org/xpl/conhome/4072023/proceeding (Conference Proceedings)

Publication series

Name	International Conference on Computational Intelligence and Security, ICCIAS

Conference

Conference	2006 International Conference on Computational Intelligence and Security, ICCIAS 2006
Country/Territory	China
City	Guangzhou
Period	3/11/06 → 6/11/06
Internet address	https://ieeexplore.ieee.org/xpl/conhome/4072023/proceeding (Conference Proceedings)

Access to Document

10.1109/ICCIAS.2006.294218

Cite this

Zhang, W., Tang, Y. Y., & Xue, Y. (2006). Handwritten Character Recognition Using Combined Gradient and Wavelet Feature. In Y. Cheung, Y. Wang, & H. Liu (Eds.), 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006 (pp. 662-667). (International Conference on Computational Intelligence and Security, ICCIAS). IEEE. https://doi.org/10.1109/ICCIAS.2006.294218

@inproceedings{99ede903aa4544e883060cc3ae011564,

title = "Handwritten Character Recognition Using Combined Gradient and Wavelet Feature",

abstract = "A novel feature based on the combination of gradient feature and coefficients of wavelet transform is developed in this paper. In handwritten character recognition, the gradient feature represents local characteristic properly, but it is sensitive to the deformation of handwritten character. Meanwhile, wavelet transform represents the character image in multiresolution analysis and keeps adequate global characteristic in different scales. In order to improve the discrimination power, we composed local and global characteristic in a combined feature. Three combination schemes are described in this paper. Experiments are conducted on two Chinese character databases, ETL8B subset (197 categories) and HKBU-SC110 (110 categories), to test the performance of proposed feature. The recognition accuracies of our feature achieve 95.53\% and 93.77\% for ETL8B subset and HKBU-SC110 by 1-NN classifier, respectively, which are higher than those of gradient feature.",

author = "Weipeng Zhang and Tang, \{Yuan Yan\} and Yun Xue",

note = "Publisher Copyright: {\textcopyright}2006 IEEE.; 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006 ; Conference date: 03-11-2006 Through 06-11-2006",

year = "2006",

month = nov,

doi = "10.1109/ICCIAS.2006.294218",

language = "English",

isbn = "1424406056",

series = "International Conference on Computational Intelligence and Security, ICCIAS",

publisher = "IEEE",

pages = "662--667",

editor = "Yiu-ming Cheung and Yuping Wang and Hailin Liu",

booktitle = "2006 International Conference on Computational Intelligence and Security, ICCIAS 2006",

address = "United States",

url = "https://ieeexplore.ieee.org/xpl/conhome/4072023/proceeding",

}

Zhang, W, Tang, YY & Xue, Y 2006, Handwritten Character Recognition Using Combined Gradient and Wavelet Feature. in Y Cheung, Y Wang & H Liu (eds), 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006. International Conference on Computational Intelligence and Security, ICCIAS, IEEE, pp. 662-667, 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006, Guangzhou, China, 3/11/06. https://doi.org/10.1109/ICCIAS.2006.294218

Handwritten Character Recognition Using Combined Gradient and Wavelet Feature. / Zhang, Weipeng; Tang, Yuan Yan; Xue, Yun.
2006 International Conference on Computational Intelligence and Security, ICCIAS 2006. ed. / Yiu-ming Cheung; Yuping Wang; Hailin Liu. IEEE, 2006. p. 662-667 (International Conference on Computational Intelligence and Security, ICCIAS).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Handwritten Character Recognition Using Combined Gradient and Wavelet Feature

AU - Zhang, Weipeng

AU - Tang, Yuan Yan

AU - Xue, Yun

PY - 2006/11

Y1 - 2006/11

N2 - A novel feature based on the combination of gradient feature and coefficients of wavelet transform is developed in this paper. In handwritten character recognition, the gradient feature represents local characteristic properly, but it is sensitive to the deformation of handwritten character. Meanwhile, wavelet transform represents the character image in multiresolution analysis and keeps adequate global characteristic in different scales. In order to improve the discrimination power, we composed local and global characteristic in a combined feature. Three combination schemes are described in this paper. Experiments are conducted on two Chinese character databases, ETL8B subset (197 categories) and HKBU-SC110 (110 categories), to test the performance of proposed feature. The recognition accuracies of our feature achieve 95.53% and 93.77% for ETL8B subset and HKBU-SC110 by 1-NN classifier, respectively, which are higher than those of gradient feature.

AB - A novel feature based on the combination of gradient feature and coefficients of wavelet transform is developed in this paper. In handwritten character recognition, the gradient feature represents local characteristic properly, but it is sensitive to the deformation of handwritten character. Meanwhile, wavelet transform represents the character image in multiresolution analysis and keeps adequate global characteristic in different scales. In order to improve the discrimination power, we composed local and global characteristic in a combined feature. Three combination schemes are described in this paper. Experiments are conducted on two Chinese character databases, ETL8B subset (197 categories) and HKBU-SC110 (110 categories), to test the performance of proposed feature. The recognition accuracies of our feature achieve 95.53% and 93.77% for ETL8B subset and HKBU-SC110 by 1-NN classifier, respectively, which are higher than those of gradient feature.

UR - https://www.scopus.com/pages/publications/38549146313

U2 - 10.1109/ICCIAS.2006.294218

DO - 10.1109/ICCIAS.2006.294218

M3 - Conference proceeding

AN - SCOPUS:38549146313

SN - 1424406056

SN - 1424406048

SN - 9781424406050

T3 - International Conference on Computational Intelligence and Security, ICCIAS

SP - 662

EP - 667

BT - 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006

A2 - Cheung, Yiu-ming

A2 - Wang, Yuping

A2 - Liu, Hailin

PB - IEEE

T2 - 2006 International Conference on Computational Intelligence and Security, ICCIAS 2006

Y2 - 3 November 2006 through 6 November 2006

ER -

Handwritten Character Recognition Using Combined Gradient and Wavelet Feature

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this