Research On Encrypted Speech Content Authentication And Tampering Recovery Method Based On Perceptual Hashing In Cloud Environment

With the explosive growth of multimedia data,cloud platforms have become the first choice for people to transmit and store multimedia data.People also concern about the privacy security and integrity of multimedia data also arise,while enjoying the convenience brought by the cloud platform.The thesis addresses the threats of speech privacy security and malicious tampering in the application of encrypted speech retrieval systems,as well as the issue of content authentication of the query results and how to recover the tampering speech.Techniques such as speech scrambling method,perceptual hashing,and the square curve fitting method are used to researched on encrypted speech content authentication and tampering recovery.The specific research work are as follows:1.In order to ensure the privacy of speech and extract the authentication digest directly from the encrypted speech,a group scrambling encryption method based on Henon mapping and a scrambling encryption plan based on Duffing mapping are proposed.The experimental results show both encryption methods can ensure the privacy and security of speech transmission and storage in the cloud.At the same time,it can directly extract the authentication digest in the encrypted speech,which improves the security of the authentication system.2.In order to realize the integrity authentication of the encrypted speech stored in the cloud,an efficient encrypted speed authentication method based on uniform subband frequency band variance and perceptual hashing is proposed.Firstly,the original speech is scrambled by Henon mapping to construct an encrypted speech library in the cloud,through extracting uniform sub-band spectrum variance of the encrypted speech and constructing a hashing sequence to generate a hashing template of the cloud.In this way,a one-to-one correspondence between the encrypted speech and the hashing sequence is built.Secondly,the authentication digest of encrypted speech is extracted according to the inquiry result.Finally,the authentication digest and the hashing sequence in the cloud are matched by the Hamming distance algorithm.The experimental results show that the proposed method has good security and efficiency,and it can directly extract the authentication digest from the encrypted speech.The authentication digest not only has good discrimination and robustness,but it can accurately locates the tampered area for malicious substitution and mute attacks.3.In order to achieve integrity authentication and tampering recovery of encrypted speech stored in the cloud,an encrypted speech authentication and tampering recovery method based on perceptual hashing and least square curve fitting is proposed.Firstly,the Duffing mapping scrambling encryption algorithm is used to encrypt the original speech to build a cloud encrypted speech library;then through extracting the product of the uniform sub-band frequency band and the spectral entropy from the encrypted speech and constructing a perceptual hashing sequence to generate a cloud hashing template.And a one-to-one correspondence between the encrypted speech and the hashing sequence is established.Scedonly,the encrypted speech authentication digest of the query result is extracted during the retrieval speech.The authentication hashing sequence and the hashing sequence in the cloud hashing template are matched using the Hamming distance algorithm,and the tampered speech signal is detected and located.Finally,the least-squares curve fitting method is used to recover the tampered speech samples.The experimental results show that the proposed scheme can extract the authentication digest directly in the encrypted speech.And the authentication digest has good discrimination and robustness.It can accurately detect and tamper with the substitution and mute attacks,and can perform high-quality recovery of tampered speech samples.