摘要 |
PROBLEM TO BE SOLVED: To make it unnecessary to execute format setting based on software in each document format, to perform character recognition without being influenced by variation in character positions of image data and to automatically search the positions of characters to be recognized. SOLUTION: A reference character string to be a reference is selected from image data of a routine document, and the coordinates of the range of the reference character string and the coordinates of an OCR character recognition target range corresponding to the reference character string are specified by a user interface of a computer system to search the image data in each page. When the reference character string can not be searched for, the range of the reference character string is moved on the image data by changing the value of the coordinates around the coordinates, a character string coincident with the reference character string on the image data is searched, and correspondingly to the movement of the range of the reference character string, the coordinates of the OCR character recognition target range corresponding to the reference character string are moved. COPYRIGHT: (C)2007,JPO&INPIT
|