Chinese character recognition by the auto recognition system

Ohhira, Takashi; Pecharanin, N.; Taguchi, Akira; Iijima, N.; Akima, Y.; Sone, M.

doi:10.1109/icnn.1995.487706

Cited by 6 publications

(8 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, in general, the current handwriting recognition technology [9][10][11] records the coordinates of pixels and the vector-value of handwriting direction in chronological order when the user writes a character. According to strokes, the technology splits them into different sequence set.…”

Section: Related Workmentioning

confidence: 99%

A Robust and On-Line Evaluation Approach to Normativity of Chinese Character Writing

Qi¹,

Huayou²,

Deng³

et al. 2010

2010 International Conference on Artificial Intelligence and Computational Intelligence

View full text Add to dashboard Cite

Chinese Character Recognition generally does its best to identify a handwritten Chinese character. It seldom cares about normativity of writings. In the paper, we propose an on-line evaluation approach to normativity of Chinese character writing, which is based on position vector of strokes of character writing. The approach can find whether or not the order and location of strokes of character writing is incorrect. It proves to be a robust and efficient method. It can be used to evaluation and teaching of Chinese characters writing.

show abstract

Section: Related Workmentioning

confidence: 99%

A Robust and On-Line Evaluation Approach to Normativity of Chinese Character Writing

Qi¹,

Huayou²,

Deng³

et al. 2010

2010 International Conference on Artificial Intelligence and Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…First, SPR can designate any MLP k . In the conventional SPR/ANN combination methods [8][9][10][11][12][13][14][15][16], a method to create ANNs corresponding to the category sets obtained from clustering has been widely used. Countermeasures for errors in clustering include manipulating the method for creating the clusters and registering several clusters in the same category.…”

Section: Architecturementioning

confidence: 99%

“…A variety of methods for combining them were attempted even before, starting with Ref. 8, and since then these methods have been evaluated [9][10][11][12][13][14][15][16]. In these attempts, the approach involved first narrowing down the candidates, and then selecting one.…”

Section: Introductionmentioning

confidence: 99%

“…In these attempts, the approach involved first narrowing down the candidates, and then selecting one. Clustering has been most often used for the first step [9,10,12,14,15], along with vector quantization [8,11], and Neural Trees [13]. Multilayer perceptrons have been most often used for the second step [8,9,11,12,15,16], along with learning vector quantization [10] and self-organizing maps [13].…”

Section: Introductionmentioning

confidence: 99%

“…In Refs. [8][9][10][11][12][13][14][15][16], the intention was to extend the types of characters to be recognized, and therefore there was an attempt to create a combination in which broad classification was performed using statistical methods, and then discrimination was performed using neural networks. However, these approaches focus on extending the types of characters, and so do not take into consideration the discrimination of similar characters, a difficulty that exists when recognizing a variety of character types.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combination of statistical and neural classifiers for a high‐accuracy recognition of large character sets

Kimura

Wakahara²,

Tomono

2005

Systems & Computers in Japan

View full text Add to dashboard Cite

SUMMARYIn this paper the authors propose a method for highaccuracy recognition of large character sets using a new combination of a statistical method and neural networks. In their method, a hierarchical structure that has several neural networks arranged in a line after the statistical method is used. First, recognition using a statistical method is performed, and this represents the final result if the top candidate does not belong to a predefined set of similar characters. If it does, then the input character is discriminated in a neural network which designates the top candidate by determining the similar characters. The results are output as final results. The basic idea of this method is the functional division of a statistical method and neural networks, and the use of a neural network as determined by a statistical method. The results of recognizing 3201 character types including JIS-1 Kanji showed an improvement in the correct recognition rate due to the combined use of a statistical method and neural networks, thereby demonstrating the validity of the authors' approach.

show abstract

Optical character recognition: An overview and an insight

Berchmans

Kumar

2014

2014 International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT)

View full text Add to dashboard Cite

Chinese character recognition by the auto recognition system

Cited by 6 publications

References 2 publications

A Robust and On-Line Evaluation Approach to Normativity of Chinese Character Writing

A Robust and On-Line Evaluation Approach to Normativity of Chinese Character Writing

Combination of statistical and neural classifiers for a high‐accuracy recognition of large character sets

Optical character recognition: An overview and an insight

Contact Info

Product

Resources

About