| Presentation | 2019-03-01 Recognition of Color Characters in Scene Images via Optimal Binarization and Distortion-tolerant Image Matching Masato Suzuki, Toru Wakahara, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper proposes an efficient combination of optimal binarization and distortion-tolerant image matching for color character recognition in scene images. First, we apply K-means clustering in Lab color space to a given color character image to generate a set of candidate binarized images. Second, those candidate binarized images are fed into the convolutional neural network in order to select an optimally binarized character image. Finally, the optimally binarized character image is recognized via affine-invariant image matching technique, GAT correlation, and k-nearest neighbor classification method. Experimental results made on the well-known color character dataset “The Chars74K” show that the proposed method achieves a correctly binarized character selection rate of 96.6%, and the final character recognition rate of 73.9%. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Character Recognition in Scene Images / Optimal Binarization / K-means Clustering / GAT Correlation |
| Paper # | PRMU2018-130,CNR2018-53 |
| Date of Issue | 2019-02-21 (PRMU, CNR) |
| Conference Information | |
| Committee | PRMU / CNR |
|---|---|
| Conference Date | 2019/2/28(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shinichi Sato(NII) / Tetsuo Ono(Hokkaido Univ.) |
| Vice Chair | Yoshihisa Ijiri(Omron) / Toru Tamaki(Hiroshima Univ.) / Masayuki Kanbara(NAIST) / Kazunori Takashio(Keio Univ.) |
| Secretary | Yoshihisa Ijiri(NEC) / Toru Tamaki(Osaka Univ.) / Masayuki Kanbara(Hokkaido Univ.) / Kazunori Takashio(Panasonic) |
| Assistant | Go Irie(NTT) / Yoshitaka Ushiku(Univ. of Tokyo) / Wataru Mito(SECOM) / Yuka Kobayashi(Toshiba) / Tatsuya Ishihara(NTT) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Cloud Network Robotics |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Recognition of Color Characters in Scene Images via Optimal Binarization and Distortion-tolerant Image Matching |
| Sub Title (in English) | |
| Keyword(1) | Character Recognition in Scene Images |
| Keyword(2) | Optimal Binarization |
| Keyword(3) | K-means Clustering |
| Keyword(4) | GAT Correlation |
| 1st Author's Name | Masato Suzuki |
| 1st Author's Affiliation | Hosei University(Hosei Univ.) |
| 2nd Author's Name | Toru Wakahara |
| 2nd Author's Affiliation | Hosei University(Hosei Univ.) |
| Date | 2019-03-01 |
| Paper # | PRMU2018-130,CNR2018-53 |
| Volume (vol) | vol.118 |
| Number (no) | PRMU-459,CNR-460 |
| Page | pp.pp.89-94(PRMU), pp.89-94(CNR), |
| #Pages | 6 |
| Date of Issue | 2019-02-21 (PRMU, CNR) |