摘要 |
<p>PROBLEM TO BE SOLVED: To provide a document electrolyzing device that can automatically generate electronic data using a marl-up language for a document, whose character recognition rate is high and which contains a chart. SOLUTION: A document picture, which has been taken in from a picture input device 11 and stored in a picture storage part 12, is displayed on a display device 14. An area is designated on the document on the display device by using a position input device 15 and a character input device 16 and attribute information is given to respective areas. A character recognition part 18 recognized characters for the respective areas by using the dictionary designated by attribute information. the result is stored in a text storage part 17b. A picture extraction part 20 extracts picture data in accordance with attribute information and stores it in a picture data storage part 17c. A mark-up part 19 executes a mark-up processing for the character area and the chart area based on attribute information and the result is stored on the text storage part.</p> |