CN-121996809-A - Method, device, electronic equipment and medium for retrieving secure outsourced image

CN121996809ACN 121996809 ACN121996809 ACN 121996809ACN-121996809-A

Abstract

The embodiment of the invention relates to the technical field of image retrieval, and provides a method, a device, electronic equipment and a medium for retrieving a safe outsourced image, which comprise the steps of obtaining an image retrieval request sent by a query user, wherein the image retrieval request comprises a retrieval encrypted image; the method comprises the steps of responding to an image retrieval request, carrying out ciphertext feature extraction on a retrieval encrypted image to obtain a first orthogonal moment feature vector, calculating a first ciphertext hash value corresponding to the first orthogonal moment feature vector, determining a target hash bucket in a hash bucket set based on the first ciphertext hash value, screening a preset number of target encrypted images with minimum distance from the first orthogonal moment feature vector in the target hash bucket, converting the encrypted form of the target encrypted image, and returning the converted target encrypted image to a query user. Therefore, efficient retrieval of the safe outsourcing images is achieved, and the safety of the scheme is improved.

Inventors

YANG TENGFEI
HE JIAWEI
LIU ZHIQUAN
QIN BAODONG
ZHANG YINGHUI
WANG TENG
MIAO MEIXIA
HUANG WENHUA

Assignees

西安邮电大学

Dates

Publication Date: 20260508
Application Date: 20251230

Claims (10)

1. A method for secure outsourcing image retrieval, comprising: acquiring an image retrieval request sent by a query user, wherein the image retrieval request carries a retrieval encryption image generated according to the query image; Performing ciphertext feature extraction on the search encrypted image in response to the image search request to obtain a first orthogonal moment feature vector corresponding to the search encrypted image, and calculating a first ciphertext hash value corresponding to the first orthogonal moment feature vector; Determining a target hash bucket which is the same as the first ciphertext hash value in a hash bucket set based on the first ciphertext hash value, wherein the hash bucket set is an image index which is constructed after encryption is performed in advance according to an original image set; Screening a preset number of target encrypted images with minimum distance from the first orthogonal moment feature vector in the target hash bucket; And converting the encrypted form of the target encrypted image, and returning the converted target encrypted image to the inquiring user.
2. The method according to claim 1, wherein the method further comprises: Acquiring an encrypted image set obtained by encrypting an original image by an image owner through an owner key; Extracting ciphertext domain features of the encrypted image set by adopting a fractional order orthogonal moment feature extraction method to obtain a second orthogonal moment feature vector, and calculating a second ciphertext hash value corresponding to the second orthogonal moment feature vector; Mapping the encrypted images with the second ciphertext hash values being equal to each other into the same hash bucket to obtain a hash bucket set; an image index is constructed based on the set of hash buckets.
3. The method of claim 1, wherein said performing ciphertext feature extraction on said retrieved encrypted image in response to said image retrieval request comprises: The method comprises the steps that ciphertext domain features of a search encrypted image are cooperatively extracted through a first cloud server and a second cloud server through a secure multiparty computing protocol, wherein the first cloud server holds a system public key and a first part of decryption keys, the second cloud server holds the system public key and a second part of decryption keys, any cloud server cannot acquire plaintext data independently, the secure multiparty computing protocol comprises a rapid cross-domain secure multiplication protocol and a rapid secure addition protocol, the rapid cross-domain secure multiplication protocol is used for realizing multiplication operation of two ciphertext data, and the rapid secure addition protocol is used for realizing addition operation of the two ciphertext data.
4. The method of claim 3, wherein the cooperatively extracting, by the first cloud server and the second cloud server, the ciphertext domain feature of the retrieved encrypted image via a secure multiparty computing protocol, comprises: The first cloud server selects four random integers, encrypts two random integers by using a query user key, and encrypts the other two random integers by using a system public key; The first cloud server carries out homomorphic operation on the ciphertext data to be multiplied of the search encrypted image and the encrypted random integer to obtain first blind ciphertext data, and after the first blind ciphertext data is decrypted for the first time by using the first partial decryption key, the first blind ciphertext data decrypted for the first time and the encrypted random integer are sent to the second cloud server, the second cloud server carries out secondary decryption by using the second partial decryption key to obtain first plaintext data with a random number mask, and the first plaintext data is re-encrypted by using the system public key to obtain first plaintext data ciphertext; And the first cloud server performs multiplication operation on the first plaintext data ciphertext to remove the influence of the random number, so as to obtain a ciphertext multiplication result.
5. The method of claim 3, wherein the cooperatively extracting, by the first cloud server and the second cloud server, the ciphertext domain feature of the retrieved encrypted image via a secure multiparty computing protocol, comprises: the first cloud server selects two random integers and encrypts the two random integers using a system public key, The first cloud server carries out homomorphic operation on the text data to be encrypted of the search encrypted image and the encrypted random integer to obtain second blind ciphertext data; The first cloud server utilizes the first partial decryption key to perform first partial decryption on the second blind ciphertext data and then sends the first blind ciphertext data to the second cloud server, the second cloud server utilizes the second partial decryption key to perform second decryption, the second decryption results are added to obtain second plaintext data with a random number mask, and the second plaintext data is re-encrypted by using the system public key to obtain second plaintext data ciphertext; And the first cloud server performs multiplication operation on the second plaintext data ciphertext to remove the influence of the random number, so as to obtain a ciphertext addition result.
6. The method according to any one of claims 3-5, wherein the performing ciphertext feature extraction on the retrieved encrypted image in response to the image retrieval request to obtain a first orthogonal moment feature vector corresponding to the retrieved encrypted image includes: And extracting ciphertext domain features of the search encrypted image based on the ciphertext multiplication result and the ciphertext addition result, and converting the ciphertext domain features into feature vectors to obtain first orthogonal moment feature vectors.
7. The method according to claim 1, wherein the method further comprises: after returning the converted target encrypted image to the inquiring user, carrying out watermark generation processing on the target encrypted image through a watermark authorization center to obtain a copyright verification certificate; And judging whether the image infringement exists or not based on the copyright verification certificate.
8. A secure outsourced image retrieval device, comprising: The acquisition module is used for acquiring an image retrieval request sent by a query user, wherein the image retrieval request carries a retrieval encryption image generated according to the query image; the feature extraction module is used for responding to the image retrieval request to carry out ciphertext feature extraction on the retrieval encrypted image, obtaining a first orthogonal moment feature vector corresponding to the retrieval encrypted image, and calculating a first ciphertext hash value corresponding to the first orthogonal moment feature vector; The retrieval module is used for determining a target hash bucket which is the same as the first ciphertext hash value in a hash bucket set based on the first ciphertext hash value, wherein the hash bucket set is an image index which is constructed after encryption is carried out in advance according to an original image set; the retrieval module is used for screening a preset number of target encrypted images with minimum distance from the first orthogonal moment feature vector in the target hash bucket; and the retrieval module is used for converting the encrypted form of the target encrypted image and returning the converted target encrypted image to the inquiring user.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the secure outsourced image retrieval method of any of claims 1 to 7 when the computer program is executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the secure outsource image retrieval method of any of claims 1 to 7.

Description

Method, device, electronic equipment and medium for retrieving secure outsourced image Technical Field The present invention relates to the field of image retrieval technologies, and in particular, to a method and apparatus for retrieving a secure outsourced image, an electronic device, and a medium. Background With the development of cloud technology, more and more image owners, including individuals and businesses, prefer to outsource large amounts of image data to remote "honest but curious" cloud servers to save local storage and computing overhead. However, due to the openness and semi-trust nature of the cloud server itself, an illegal entity including the cloud server itself may steal sensitive information contained in the image, such as personal identity, home address, work unit, and the like. To protect the privacy of images and to enable image sharing. Secure outsourced image retrieval is widely studied. However, the existing secure outsourced image retrieval system adopts a shared key and does not support key revocation, so that the image owner and the query user cannot be dynamically deleted, the expandability of the system for dynamically updating the image owner and the query user is weak, and the actual scene of multiple sources and multiple users cannot be met. Moreover, once an illegal operation occurs, an illegal user who cannot revoke the owned key will bring a great loss to a legal user. Furthermore, the index of the secure outsourced image retrieval system is built by the image owner, and once the image owner needs to upload or cancel cloud image data, the image owner needs to spend huge communication overhead and calculation overhead to update the index, so that the system has weak expandability of dynamically updating the image data. In addition, after the query user decrypts the retrieved ciphertext image to obtain the plaintext image, illegal secondary distribution of the plaintext image may be performed, which results in risk of image infringement. The prior art adopts encryption domain information hiding technology to damage the content of the original image. Therefore, how to realize copyright protection of images and tracking of illegal inquiry users on the premise of guaranteeing the integrity of image data is a problem to be solved urgently. Disclosure of Invention The invention provides a method, a device, electronic equipment and a medium for searching a safe outsourcing image, which are used for solving the defects that the content of an original image is damaged and the secret key sharing safety is low by adopting an encryption domain information hiding technology in the prior art, realizing efficient searching, copyright protection and illegal inquiring user tracking of the safe outsourcing image and improving the safety of a scheme. The invention provides a safe outsourcing image retrieval method, which comprises the following steps: acquiring an image retrieval request sent by a query user, wherein the image retrieval request carries a retrieval encryption image generated according to the query image; Performing ciphertext feature extraction on the search encrypted image in response to the image search request to obtain a first orthogonal moment feature vector corresponding to the search encrypted image, and calculating a first ciphertext hash value corresponding to the first orthogonal moment feature vector; Determining a target hash bucket which is the same as the first ciphertext hash value in a hash bucket set based on the first ciphertext hash value, wherein the hash bucket set is an image index which is constructed after encryption is performed in advance according to an original image set; Screening a preset number of target encrypted images with minimum distance from the first orthogonal moment feature vector in the target hash bucket; And converting the encrypted form of the target encrypted image, and returning the converted target encrypted image to the inquiring user. In one possible embodiment, the method further comprises: Acquiring an encrypted image set obtained by encrypting an original image by an image owner through an owner key; Extracting ciphertext domain features of the encrypted image set by adopting a fractional order orthogonal moment feature extraction method to obtain a second orthogonal moment feature vector, and calculating a second ciphertext hash value corresponding to the second orthogonal moment feature vector; Mapping the encrypted images with the second ciphertext hash values being equal to each other into the same hash bucket to obtain a hash bucket set; an image index is constructed based on the set of hash buckets. In one possible embodiment, the method further comprises: The method comprises the steps that ciphertext domain features of a search encrypted image are cooperatively extracted through a first cloud server and a second cloud server through a secure multiparty computing protocol, wherein the first cloud server holds a sys