Method for adding and extracting watermark in audio signal

Abstract

The invention discloses a method for adding and extracting watermark in an audio signal. The one-dimensional audio signal is converted to a two-dimensional audio image, a random template is generated, diffraction optical transformation of the audio image, watermark information, and the random template is performed based on an optical diffraction imaging theory, then weighted stacking is performed, and information hiding of the watermark information in the audio image is realized. By employing the advantages of multiple dimensions and large capacity of optical diffraction imaging, hiding of the large-capacity watermark information in audio information is realized, and problems that according to the conventional method, the embedding amount of the watermark information is small, and information hiding is greatly limited are solved; by employing the characteristic of the tearing property of optical diffraction imaging, the function of extracting complete watermark information by employing a part of incomplete audio images is realized; authentication and certification of the reliability of a source of the audio information can be realized through extraction of the watermark information; and an original carrier image is not needed during watermark extraction, and blind extraction of the watermark information is realized.

Description

technical field [0001] The invention belongs to the field of information security, and in particular relates to a method for adding and extracting watermarks in audio signals. Background technique [0002] With the popularity of digital multimedia information applications and modern equipment, voice (audio signal) has become more and more easy to obtain, and is gradually mastered by ordinary people. Voice also has its important characteristics of being authentic and difficult to imitate. Therefore, the scope of application of voice is gradually expanding. For example, recording materials have become an important means of judicial certification and gradually become an important basis for judicial judgment. Article 70 of the "Several Provisions of the Supreme People's Court Concerning Evidence in Civil Procedures" promulgated in 2002 stipulates: "Where one party raises the following evidence, and the other party raises an objection but there is no sufficient evidence to the co...

AI Technical Summary

Problems solved by technology

Article 70 of the "Several Provisions of the Supreme People's Court Concerning Evidence in Civil Procedures" promulgated in 2002 stipulates: "Where one party raises the following evidence, and the other party raises an objection but there is no sufficient evidence to the contrary, the people's court shall confirm the proof Force: ... (3) Undoubted audio-visual materials or copies of audio-visual materials that have been verified by other evidence and obtained by legal means; ..." However, Article 68 of this provision also clarifies that "the Evidence obtained by infringing upon the lawful rights and interests of others or violating the prohibitive provisions of the law cannot be used as the basis for ascertaining the facts of the case

This is developed with the help of existing image information hiding technology. These methods often hide less information and are very limited. However, these algorithms currently do not consider the authentication and notarization of information sources.

Images