On distributed representation of output layer for recognizing Japanese Kana characters using neural networks

On distributed representation of output layer for recognizing Japanese Kana characters using neural networks

フォーマット:

論文

責任表示:

Keeni, Kanad ; Shimodaira, Hiroshi ; Nakayama, Kenji

言語:

英語

出版情報:

IEEE(Institute of Electrical and Electronics Engineers), 1997-08-01

著者名:

掲載情報:

Proceedings of the International Conference on Document Analysis and Recognition, ICDAR

巻:

2

開始ページ:

600

終了ページ:

603

バージョン:

publisher

概要:

This paper presents an automatic coding scheme for representing the output layer of a neural network. Compared to local representation where the number of output unit is p, the number of output unit required for the proposed representation is close to log p. The output of seven different printers were used for evaluating the performance of the system. The proposed automatic representation gave the average recognition rate of 98.7% for 71 categories.

URL:

http://hdl.handle.net/2297/6795

タイトル・著者・出版者が同じ資料

詳細

主題:	Character recognition; Neural networks; Output layer representation; Pattern recognition systems
受理日:	1997-08-01

類似資料:

1

2

3

4

5

6

7

8

9

10

11

12

1 論文 Automatic generation of initial weights and target outputs of multilayer neural networks and its application to pattern classification Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi International Conference on Neural Information Processing (ICONIP) / Gazelle Distribution Trade Gxc	7 論文 Probabilistic memory capacity of recurrent neural networks Miyoshi, Seiji, Nakayama, Kenji IEEE(Institute of Electrical and Electronics Engineers)
2 論文 Estimation of initial weights and hidden units for fast learning of multi-layer neural networks for pattern classification Keeni, Kanad, Nakayama, Kenji, Shimodaira, Hiroshi IEEE(Institute of Electrical and Electronics Engineers)	8 論文 Optimization of activation functions in multilayer neural network applied to pattern classification Nakayama, Kenji, Kimura, Yoshinori IEEE(Institute of Electrical and Electronics Engineers)
3 論文 Quantization level increase in human face images using multilayer neural network Nakayama, Kenji, Kimura, Yoshinori, Katayama, Hiroshi IEEE(Institute of Electrical and Electronics Engineers)	9 論文 Japanese Kanji Character Recognition Using Cellular Neural Networks and Modified Self-Organizing Feature Map Nakayama, Kenji, Chigawa, Y. Institute of Electrical and Electronics Engineers (IEEE)
4 論文 Signal classification based on frequency analysis using multilayer neural network with limited data and computations Hara, Kazuyuki, Nakayama, Kenji IEEE(Institute of Electrical and Electronics Engineers)	10 論文 A neural demodulator for amplitude shift keying signals Nakayama, Kenji, Imai, Kunihiko IEEE(Institute of Electrical and Electronics Engineers)
5 論文 Memory capacity bound and threshold optimization in recurrent neural network with variable hysteresis threshold Nakayama, Kenji, Nishimura, Katsuaki, Katayama, Hiroshi IEEE(Institute of Electrical and Electronics Engineers)	11 論文 A delta rule algorithm using double hysteresis thresholds for recurrent associative memory Nakayama, Kenji, Nishimura, Katsuaki IEEE(Institute of Electrical and Electronics Engineers)
6 論文 A structure trainable neural network with embedded gate units: Multi-dimensional input/output and its learning algorithm Nakayama, Kenji, Hirano, Akihiro, Kourin, Makoto IEEE(Institute of Electrical and Electronics Engineers)	12 論文 A hybrid multilayer neural network for binary pattern classification and its low-bit learning algorithm Nakayama, Kenji, Katayama, Hiroshi IEEE(Institute of Electrical and Electronics Engineers)