| Presentation | 2008-02-21 Word Extraction Method by Generating Multiple Character Hypotheses Hiroaki TAKEBE, Katsuhito FUJIMOTO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | It is necessary to extract words which are headers and data for recognizing logical structure of form images. However, word extraction fails if the part of characters is not extracted because of layout analysis error or character recognition error. Words of data usually consist of numbers or symbols which are similar to many characters, so they are often mis-extracted due to character recognition error. Moreover, words of data are represented by regular expression which includes variable numbers of wild-card character. We propose the word extraction method which generates multiple character hypotheses by controlling recognition target, and searches the combination of recognition result with character of regular expression by lattice matching based on the local optimum. It was confirmed that the proposed method improved extraction accuracy of header and data by the experiment for the real form images. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Form Image / Word / Header / Data / Recognition Error / Regular Expression / Lattice |
| Paper # | PRMU2007-217 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2008/2/14(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) | Word Extraction Method by Generating Multiple Character Hypotheses |
| Sub Title (in English) | |
| Keyword(1) | Form Image |
| Keyword(2) | Word |
| Keyword(3) | Header |
| Keyword(4) | Data |
| Keyword(5) | Recognition Error |
| Keyword(6) | Regular Expression |
| Keyword(7) | Lattice |
| 1st Author's Name | Hiroaki TAKEBE |
| 1st Author's Affiliation | FUJITSU LABORATORIES LTD.() |
| 2nd Author's Name | Katsuhito FUJIMOTO |
| 2nd Author's Affiliation | FUJITSU LABORATORIES LTD. |
| Date | 2008-02-21 |
| Paper # | PRMU2007-217 |
| Volume (vol) | vol.107 |
| Number (no) | 491 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |