| Presentation | 2007-03-15 An Analytic Word Recognition Algorithm Using a Posteriori Probability : Normalization of the number of segmentation candidates Tomoyuki HAMAMURA, Takuma AKAGI, Bunpei IRIE, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Word recognition algorithms are classified into two major groups. One is an "analytic" approach of recognizing individual characters, while the other is a "holistic" approach dealing with an entire word image. In the former approach, we have proposed an evaluation function (Normalized a posteriori probability ratio) that has been derived from an a posteriori probability of a word in a lexicon in order to make it easy-to-compute. In this paper, we point out that an approximation error of this evaluation function be fairly large when an employed segmentation method generates many segments. Furthermore, improved derivation of the a posteriori probability is proposed, and a new evaluation function (enhanced a posteriori probability ratio) is derived, which is more precise without losing easy-to-compute feature. A result of an experiment using real images shows 9.1 % improvement on handwritten word recognition. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Word recognition / Analytic / Evaluation function / A posteriori probability / A posteriori probability ratio / Normalized a posteriori probability ratio / Enhanced a posteriori probability ratio |
| Paper # | PRMU2006-238 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2007/3/8(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) | An Analytic Word Recognition Algorithm Using a Posteriori Probability : Normalization of the number of segmentation candidates |
| Sub Title (in English) | |
| Keyword(1) | Word recognition |
| Keyword(2) | Analytic |
| Keyword(3) | Evaluation function |
| Keyword(4) | A posteriori probability |
| Keyword(5) | A posteriori probability ratio |
| Keyword(6) | Normalized a posteriori probability ratio |
| Keyword(7) | Enhanced a posteriori probability ratio |
| 1st Author's Name | Tomoyuki HAMAMURA |
| 1st Author's Affiliation | TOSHIBA Corp.() |
| 2nd Author's Name | Takuma AKAGI |
| 2nd Author's Affiliation | TOSHIBA Corp. |
| 3rd Author's Name | Bunpei IRIE |
| 3rd Author's Affiliation | TOSHIBA Corp. |
| Date | 2007-03-15 |
| Paper # | PRMU2006-238 |
| Volume (vol) | vol.106 |
| Number (no) | 605 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |