In this paper, a new idea of digital document authentication is proposed. We deal with this problem by treating a digital document as an image, hence the authentication code of a digital document can be derived from this document’s text itself. No longer to encrypt a document by heavily enlarging the document size with a given complex digital signature, the authentication code is just encoded into the original document image version.
The most remarkable benefit of our method is the document can be maintained at the same size after authentication code encryption. Moreover, by adopting this approach, not only the tampered document can be verified by itself, but also the tampered words are able to be located precisely. In the experiment, some attacks such as words substitutions have been conducted to show the performance of our method.