This paper describes an adaptive feature extraction method that exploits category-specific information to overcome both image degradation and deformation in character recognition. When recognizing multiple fonts, geometric features such as directional information of strokes are often used but they are weak against the deformation and degradation that appear in videos or natural scenes. To tackle these problems, the proposed method estimates the degree of deformation and degradation of an input pattern by comparing the input pattern and the template of each category as category-specific information. This estimation enables us to compensate the aspect ratio associated with shape and the degradation in feature values and so obtain higher recognition accuracy. Recognition experiments using characters extracted from videos show that the proposed method is superior to the conventional alternatives in resisting deformation and degradation.
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
Minoru MORI, Minako SAWAKI, Junji YAMATO, "Robust Character Recognition Using Adaptive Feature Extraction Method" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 1, pp. 125-133, January 2010, doi: 10.1587/transinf.E93.D.125.
Abstract: This paper describes an adaptive feature extraction method that exploits category-specific information to overcome both image degradation and deformation in character recognition. When recognizing multiple fonts, geometric features such as directional information of strokes are often used but they are weak against the deformation and degradation that appear in videos or natural scenes. To tackle these problems, the proposed method estimates the degree of deformation and degradation of an input pattern by comparing the input pattern and the template of each category as category-specific information. This estimation enables us to compensate the aspect ratio associated with shape and the degradation in feature values and so obtain higher recognition accuracy. Recognition experiments using characters extracted from videos show that the proposed method is superior to the conventional alternatives in resisting deformation and degradation.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.125/_p
Copy
@ARTICLE{e93-d_1_125,
author={Minoru MORI, Minako SAWAKI, Junji YAMATO, },
journal={IEICE TRANSACTIONS on Information},
title={Robust Character Recognition Using Adaptive Feature Extraction Method},
year={2010},
volume={E93-D},
number={1},
pages={125-133},
abstract={This paper describes an adaptive feature extraction method that exploits category-specific information to overcome both image degradation and deformation in character recognition. When recognizing multiple fonts, geometric features such as directional information of strokes are often used but they are weak against the deformation and degradation that appear in videos or natural scenes. To tackle these problems, the proposed method estimates the degree of deformation and degradation of an input pattern by comparing the input pattern and the template of each category as category-specific information. This estimation enables us to compensate the aspect ratio associated with shape and the degradation in feature values and so obtain higher recognition accuracy. Recognition experiments using characters extracted from videos show that the proposed method is superior to the conventional alternatives in resisting deformation and degradation.},
keywords={},
doi={10.1587/transinf.E93.D.125},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - Robust Character Recognition Using Adaptive Feature Extraction Method
T2 - IEICE TRANSACTIONS on Information
SP - 125
EP - 133
AU - Minoru MORI
AU - Minako SAWAKI
AU - Junji YAMATO
PY - 2010
DO - 10.1587/transinf.E93.D.125
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2010
AB - This paper describes an adaptive feature extraction method that exploits category-specific information to overcome both image degradation and deformation in character recognition. When recognizing multiple fonts, geometric features such as directional information of strokes are often used but they are weak against the deformation and degradation that appear in videos or natural scenes. To tackle these problems, the proposed method estimates the degree of deformation and degradation of an input pattern by comparing the input pattern and the template of each category as category-specific information. This estimation enables us to compensate the aspect ratio associated with shape and the degradation in feature values and so obtain higher recognition accuracy. Recognition experiments using characters extracted from videos show that the proposed method is superior to the conventional alternatives in resisting deformation and degradation.
ER -