| Presentation | 2009-01-29 An Improved Algorithm to Generate Feature Graphs for Handwritten Character Recognition Masaki HAYASHI, Shuichi NISHIDA, Mariko KURAMOCHI, Asami NINOMIYA, Mitsuru NAKATA, Qi-Wei GE, Makoto YOSHIMURA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, we propose an improved algorithm to generate feature graphs for handwritten character recognition. A feature graph is a graph representing the structures of characters, and character fitness is an index showing similarity of feature graphs. We have proposed a method of handwritten character recognition by character fitness. However, our previous algorithm had some problems: 1) vertices representing a curve (called curve vertex) are generated more than necessary; 2) 4-degree vertices representing crosses are divided into pairs of 3-degree vertices. These problems lower the precision of our character recognition. In this paper, to solve these problems, we improve our previous algorithm by modifying conditions of inflection point process and adding the second-clustering process. And further we evaluate our new algorithm by comparing the results new algorithm and the previous one. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | handwritten character recognition / feature graph / character fitness / clustering / thinning |
| Paper # | CST2008-48 |
| Date of Issue |
| Conference Information | |
| Committee | CST |
|---|---|
| Conference Date | 2009/1/22(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 | Concurrent System Technology (CST) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | An Improved Algorithm to Generate Feature Graphs for Handwritten Character Recognition |
| Sub Title (in English) | |
| Keyword(1) | handwritten character recognition |
| Keyword(2) | feature graph |
| Keyword(3) | character fitness |
| Keyword(4) | clustering |
| Keyword(5) | thinning |
| 1st Author's Name | Masaki HAYASHI |
| 1st Author's Affiliation | Graduate School of Education, Yamaguchi University() |
| 2nd Author's Name | Shuichi NISHIDA |
| 2nd Author's Affiliation | Graduate School of Education, Yamaguchi University |
| 3rd Author's Name | Mariko KURAMOCHI |
| 3rd Author's Affiliation | Faculty of Education, Yamaguchi University |
| 4th Author's Name | Asami NINOMIYA |
| 4th Author's Affiliation | Faculty of Education, Yamaguchi University |
| 5th Author's Name | Mitsuru NAKATA |
| 5th Author's Affiliation | Faculty of Education, Yamaguchi University |
| 6th Author's Name | Qi-Wei GE |
| 6th Author's Affiliation | Faculty of Education, Yamaguchi University |
| 7th Author's Name | Makoto YOSHIMURA |
| 7th Author's Affiliation | Faculty of Education, Yamaguchi University |
| Date | 2009-01-29 |
| Paper # | CST2008-48 |
| Volume (vol) | vol.108 |
| Number (no) | 415 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |