| 摘要 |
PROBLEM TO BE SOLVED: To make it possible to read entry items entered in respective documents distinctively and accurately, by detecting ID information specifying the document kind in text data obtained by converting read image data. SOLUTION: In an ID information memory, a document title 3 is stored in the form of an ID character string which is entered at an arbitrary position in each document and specifies the kind of the document. Further, a character position table 5 contains the respective character positions in text data of respective entry items of documents 1 corresponding to ID numbers specifying the kinds of the documents by the ID numbers. Then, image data of a document 1 are read by an image read part 6 and converted into text data by a character conversion part 7. The document 3 is detected from the text data and the kind of the document 1 is specified. The character positions in the text data of the respective entry items of the document are read out of the character position table 5 and the characters at the text positions in the text data are read as corresponding entry items. |
