Research On Personal Recognition Approach Of Network Based On Combined Digital Watermark With Palmprint

Authentication is the first barrier of the information system security field. It is a key field in the information age. Identity authentication has become the basic service for its application with the fast progressing of Internet and information going, which the identity authentication has got new requirement to discard the old way of authentication and design the integrative authentication system, which also is the current of identity authentication technology.With the fast development of electronics, information, computer and networks, biometrics has drawn more and more attention. As one of the most excellent identifying methods in biometrics, palmprint identification's theory and application appeal a lot of interest. Biometrics characterized by palmprint is an increasing development trend of user authentication, and has promising application in future.But at present identification document is still used popularly and widely, though it has less and less security affront more and more attack. It makes enormous losing for the country and society. So it is necessary enough to protect the identification document. Then the technology of digital watermarking provides a new way to solve the problem. Digital watermarking as a new mean to provide copyright protection and authentication of multimedia data has been developing rapidly since 1990's, and becomes an internationally hot topic at present. A digital watermarking technology allows users to imperceptibly embed a specifically signal (the watermark) into multimedia data such as image, text, video and audio by a certain proposed algorithm in order to protect multimedia data and authentication.Customarily, the watermark signal is number, text or picture etc. This kind of watermark exists two problems: one is the uniqueness of the watermark; another is that there is no inevitable contact between the watermark and the owner of the paper.This paper presents a new approach of remote network verification. The model combine digital watermark with handprint recognition, by imbedding fragile watermark to recognize the handprint has been destroyed whether or not. And the watermarked handprint is hidden in the general public image to transmit secretly in case of attack the attackers'attention. We combine the wavelet moments with BP neural network to finish the supervised personal identification.