| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a technology for increasing the probability of extracting correct watermark information even when the watermark information is not repeatedly embedded and even when the number of character intervals is changed compared to the time of embedding. Ž<P>SOLUTION: The character interval in a document image is acquired in a document information acquisition unit 702, and a data part is extracted using the size relation of the character interval in a data part extraction unit 703. Whether or not the number of the character intervals of the data part is changed from the time of embedding is determined in a determination unit 704, and when determining that the number of the character intervals of the data part is changed from the time of embedding, the eliminated bit of data is specified in an eliminated bit specifying unit 705, and error correction encoded data are generated on the basis of the eliminated bit and the character intervals in a generation unit 706. In a watermark information extraction unit 707, the error correction encoded data are decoded and the watermark information is extracted. Ž<P>COPYRIGHT: (C)2010,JPO&INPIT Ž
|
