The performance of integrated segmentation and recognition of handwritten numeral strings relies on the classification accuracy and the non-character resistance of the underlying character classifier, which is variable depending on the techniques of pattern normalization, feature extraction, and classifier structure. In this paper, we evaluate the effects of 12 normalization functions and four selected feature types on numeral string recognition. Slant correction (deslant) is combined with the normalization functions and features so as to create 96 feature vectors, which are classified using two classifier structures. In experiments on numeral string images of the NIST Special Database 19, the classifiers have yielded very high string recognition accuracies. We show the superiority of moment normalization with adaptive aspect ratio mapping and the gradient direction feature, and observed that slant correction is beneficial to string recognition when combined with good normalization methods.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Cheng-Lin LIU, Hiroshi SAKO, Hiromichi FUJISAWA, "Handwritten Numeral String Recognition: Effects of Character Normalization and Feature Extraction" in IEICE TRANSACTIONS on Information,
vol. E88-D, no. 8, pp. 1791-1798, August 2005, doi: 10.1093/ietisy/e88-d.8.1791.
Abstract: The performance of integrated segmentation and recognition of handwritten numeral strings relies on the classification accuracy and the non-character resistance of the underlying character classifier, which is variable depending on the techniques of pattern normalization, feature extraction, and classifier structure. In this paper, we evaluate the effects of 12 normalization functions and four selected feature types on numeral string recognition. Slant correction (deslant) is combined with the normalization functions and features so as to create 96 feature vectors, which are classified using two classifier structures. In experiments on numeral string images of the NIST Special Database 19, the classifiers have yielded very high string recognition accuracies. We show the superiority of moment normalization with adaptive aspect ratio mapping and the gradient direction feature, and observed that slant correction is beneficial to string recognition when combined with good normalization methods.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.8.1791/_p
Copy
@ARTICLE{e88-d_8_1791,
author={Cheng-Lin LIU, Hiroshi SAKO, Hiromichi FUJISAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Handwritten Numeral String Recognition: Effects of Character Normalization and Feature Extraction},
year={2005},
volume={E88-D},
number={8},
pages={1791-1798},
abstract={The performance of integrated segmentation and recognition of handwritten numeral strings relies on the classification accuracy and the non-character resistance of the underlying character classifier, which is variable depending on the techniques of pattern normalization, feature extraction, and classifier structure. In this paper, we evaluate the effects of 12 normalization functions and four selected feature types on numeral string recognition. Slant correction (deslant) is combined with the normalization functions and features so as to create 96 feature vectors, which are classified using two classifier structures. In experiments on numeral string images of the NIST Special Database 19, the classifiers have yielded very high string recognition accuracies. We show the superiority of moment normalization with adaptive aspect ratio mapping and the gradient direction feature, and observed that slant correction is beneficial to string recognition when combined with good normalization methods.},
keywords={},
doi={10.1093/ietisy/e88-d.8.1791},
ISSN={},
month={August},}
Copy
TY - JOUR
TI - Handwritten Numeral String Recognition: Effects of Character Normalization and Feature Extraction
T2 - IEICE TRANSACTIONS on Information
SP - 1791
EP - 1798
AU - Cheng-Lin LIU
AU - Hiroshi SAKO
AU - Hiromichi FUJISAWA
PY - 2005
DO - 10.1093/ietisy/e88-d.8.1791
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2005
AB - The performance of integrated segmentation and recognition of handwritten numeral strings relies on the classification accuracy and the non-character resistance of the underlying character classifier, which is variable depending on the techniques of pattern normalization, feature extraction, and classifier structure. In this paper, we evaluate the effects of 12 normalization functions and four selected feature types on numeral string recognition. Slant correction (deslant) is combined with the normalization functions and features so as to create 96 feature vectors, which are classified using two classifier structures. In experiments on numeral string images of the NIST Special Database 19, the classifiers have yielded very high string recognition accuracies. We show the superiority of moment normalization with adaptive aspect ratio mapping and the gradient direction feature, and observed that slant correction is beneficial to string recognition when combined with good normalization methods.
ER -