| Presentation | 2014-03-07 Rotation-robust Feature Extraction for Image Recognition Yuya IWASAKI, JAEHOON yu, Ryusuke MIYAMOTO, Takao ONOYE, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In the field of the optical character recognition, to recognize characters captured via camera in natural images has become a new trend as humans can realize such characters. Unlike the recognition of printed characters, schemes for character recognition in natural images must treat several disturbance that often happens when natural images are captured by cameras. In this paper, we propose a novel feature extraction scheme that can achieve accurate classification even if targets are heavily rotated at capturing process. Experimental results using the MNIST dataset show that the proposed scheme can achieve 85.34% classification if the feature extraction scheme is invariant with respect to 360 degrees rotation. If the tolerance to rotation of the proposed scheme is limited, the recognition accuracy reaches more than 90% when the range of rotation applied to input characters is more than -30 degrees and less than 30 degrees. Moreover, the proposed scheme with the limited tolerance to rotation can recognize 96.67% of input characters when the input characters are not rotated as accurate as the other existing schemes. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | feature extraction / rotation-robust / image recognition / HOG / optical character recognition |
| Paper # | SIS2013-68 |
| Date of Issue |
| Conference Information | |
| Committee | SIS |
|---|---|
| Conference Date | 2014/2/27(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 | Smart Info-Media Systems (SIS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Rotation-robust Feature Extraction for Image Recognition |
| Sub Title (in English) | |
| Keyword(1) | feature extraction |
| Keyword(2) | rotation-robust |
| Keyword(3) | image recognition |
| Keyword(4) | HOG |
| Keyword(5) | optical character recognition |
| 1st Author's Name | Yuya IWASAKI |
| 1st Author's Affiliation | Dept. of Information Systems Engineering, Graduate School of Information Science and Technology, Osaka University() |
| 2nd Author's Name | JAEHOON yu |
| 2nd Author's Affiliation | Dept. of Information Systems Engineering, Graduate School of Information Science and Technology, Osaka University |
| 3rd Author's Name | Ryusuke MIYAMOTO |
| 3rd Author's Affiliation | Dept. of Information Science School of Science and Technology |
| 4th Author's Name | Takao ONOYE |
| 4th Author's Affiliation | Dept. of Information Systems Engineering, Graduate School of Information Science and Technology, Osaka University |
| Date | 2014-03-07 |
| Paper # | SIS2013-68 |
| Volume (vol) | vol.113 |
| Number (no) | 467 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |