| Presentation | 1998/2/19 Accuracy Improvement by Gradient feature and Variable Absorbable Covariance Matrix in Handwritten Chinese Character Recognition Tetsushi WAKABAYASHI, Kazuhiro SAWA, Shinji TSURUOKA, Fumitaka KIMURA, Yasuji MIYAKE, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this report, a procedure of gradient feature extraction for handwritten Chinese character recognition and enhanced learning method by variable absorbable covariance matrix are proposed, and the efficiency is evaluated by Chinese character recognition experiments. In the gradient feature extraction, it it shown that the refinement of the weighting filter and the magnification of the character image improve the recognition accuracy. Variable absorbable covariance matrix is obtained from sample covariance matrix by adding error variance, such as displacement and inclination, which are known a priori to be common to all classes. This is aimed to enhance the learning capability to handle the test pattern outside the range of the training samples. A high recognition rate of 99.41% was achieved for all test samples of ETL9B by the proposed techniques. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Chinese character recognition / feature extraction / gradient / ETL9B / variable absorbable auto-correlation matrix / covariance matrix |
| Paper # | |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 1998/2/19(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Pattern Recognition and Media Understanding (PRMU) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Accuracy Improvement by Gradient feature and Variable Absorbable Covariance Matrix in Handwritten Chinese Character Recognition |
| Sub Title (in English) | |
| Keyword(1) | Chinese character recognition |
| Keyword(2) | feature extraction |
| Keyword(3) | gradient |
| Keyword(4) | ETL9B |
| Keyword(5) | variable absorbable auto-correlation matrix |
| Keyword(6) | covariance matrix |
| 1st Author's Name | Tetsushi WAKABAYASHI |
| 1st Author's Affiliation | Department of Information Engineering, Faculty of Engineering, Mie University() |
| 2nd Author's Name | Kazuhiro SAWA |
| 2nd Author's Affiliation | Department of Information Engineering, Faculty of Engineering, Mie University |
| 3rd Author's Name | Shinji TSURUOKA |
| 3rd Author's Affiliation | Department of Electrical and Electronic Engineering, Faculty of Engineering, Mie University |
| 4th Author's Name | Fumitaka KIMURA |
| 4th Author's Affiliation | Department of Information Engineering, Faculty of Engineering, Mie University |
| 5th Author's Name | Yasuji MIYAKE |
| 5th Author's Affiliation | Department of Information Engineering, Faculty of Engineering, Mie University |
| Date | 1998/2/19 |
| Paper # | |
| Volume (vol) | vol.97 |
| Number (no) | 558 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |