Block chain assisted privacy protection image retrieval method
A technology for privacy protection and image retrieval, which is applied in still image data retrieval, still image data indexing, and digital data protection. Solve the problem of data integrity verification and ensure the effect of security
Pending Publication Date: 2022-06-03
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Existing work on image retrieval mostly focuses on a single semi-trusted cloud server, and research on multi-cloud image retrieval is very limited
In addition, there are two security issues that need to be addressed in multi-cloud image retrieval
On the one hand, because the multi-cloud central retrieval server may return false or incomplete search results in order to obtain commercial profits, thereby reducing the accuracy of retrieval, which seriously damages the interests of users
For example, a user pays and wants to download a similar image from a cloud server, but the retrieved image is far from what he / she wants
On the other hand, in order to save computing and bandwidth resources, the cloud server may delete or tamper with images that are not frequently accessed by users, thereby destroying the integrity of image data, resulting in the loss of important information of the image owner, or errors in the decryption of images by users
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0045]The present invention will be further described in detail below in conjunction with the accompanying drawings.
[0046] The present invention proposes a blockchain-assisted privacy-preserving image retrieval method, such as figure 1 shown, including the following steps:
[0047] Data is processed before being uploaded to the cloud. The process is as follows:
[0048] First, the image owner uses the convolutional neural network model to extract image features; and then uses the kNN algorithm to encrypt each image feature. The specific process is as follows:
[0049] for each image I i The eigenvector F of i =(F i1 ,F i2 ,…,F in ), the image owner expands F in turn i for F i '=(F i1 ,F i2 ,…,F in ,||F i || 2 ) T , then put F i ' into two parts F i 1 and F i 2 , calculated as follows:
[0050]
[0051] Among them, S j ∈S, F ij ∈(F i1 ,F i2 ,…,F in ), S is a random bit string, v is a random value, F i is image I i eigenvector of , F i1 ,F i2 ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a blockchain-assisted privacy protection image retrieval method, which comprises the following steps: firstly, an image owner extracts image features by using a CNN model, then encrypts an image and the features by using an SM4 algorithm and a secure kNN algorithm, and uploads a ciphertext image to a cloud server; then, the image owner calculates the hash of the ciphertext image and generates a Merkle hash tree, then an index table of encryption features is constructed by using an LSH algorithm, and the index table and the Merkle tree are uploaded to an alliance chain; and finally, the user uploads the encryption feature of the query image to the alliance chain, retrieves the IDs of the first k images similar to the image, and returns the k IDs to the cloud, the cloud sends the corresponding image to the user, and the user decrypts the image to obtain a plaintext image. The advanced block chain technology and the cryptography technology are adopted, the accuracy of the retrieval result can be guaranteed, and the integrity of the image data stored in the cloud can be verified; meanwhile, the image retrieval problem in the multi-cloud environment is solved, and the efficiency of the retrieval scheme is improved.
Description
technical field [0001] The invention belongs to the technical field of image processing and data auditing, and in particular relates to a blockchain-assisted privacy protection image retrieval method. Background technique [0002] With the increasing popularity of mobile devices, a large number of high-resolution images are generated every day, including in publishing, education, medicine, and more. Due to the limited storage space of electronic devices, most image owners choose to store their images in cloud services, such as Apple Cloud, Baidu Cloud, Tencent Cloud, Alibaba Cloud, etc. Facing large image datasets, efficient image retrieval becomes a common requirement. Currently, there are two main techniques for image retrieval: text-based retrieval and content-based retrieval. Text-based image retrieval methods need to generate text descriptors for each image, so there are some shortcomings. First, the accuracy of retrieval results completely depends on the accuracy of...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06F16/583G06F16/51G06F21/64
CPCG06F16/583G06F16/51G06F21/64Y02D10/00
Inventor 田苗张玉书
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS