| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2010: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2009: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2008: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2007: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Research Abstract |
We have proposed a character recognition method for consecutive handwritten characters written in Japanese historical documents and implemented a prototype recognition program based on this method. In our method, structures of the characters are represented by feature graphs and dictionary storing known characters is prepared for character recognition. The similarities between feature graphs of target characters and dictionary characters are calculated, and the target character is recognized as the dictionary character with the highest similarity. The computational experiment using our proposed method has been done for the characters written in The Tale of Genji. As the result, we have the recognition rates 76.6% for non-consecutive target characters and 54% for consecutive target characters.
|
