A Privacy-Preserving Image Retrieval Based on AC-Coefficients and Color Histograms in Cloud Environment

Abstract

Content based image retrieval (CBIR) techniques have been widely deployed in many applications for seeking the abundant information existed in images. Due to large amounts of storage and computational requirements of CBIR, outsourcing image search work to the cloud provider becomes a very attractive option for many owners with small devices. However, owing to the private content contained in images, directly outsourcing retrieval work to the cloud provider apparently bring about privacy problem, so the images should be protected carefully before outsourcing. This paper presents a secure retrieval scheme for the encrypted images in the YUV color space. With this scheme, the discrete cosine transform (DCT) is performed on the Y component. The resulting DC coefficients are encrypted with stream cipher technology and the resulting AC coefficients as well as other two color components are encrypted with value permutation and position scrambling. Then the image owner transmits the encrypted images to the cloud server. When receiving a query trapdoor form on query user, the server extracts AC-coefficients histogram from the encrypted Y component and extracts two color histograms from the other two color components. The similarity between query trapdoor and database image is measured by calculating the Manhattan distance of their respective histograms. Finally, the encrypted images closest to the query image are returned to the query user. ]. Nevertheless, huge storage space and complex computation requirements are needed to search one particular image from a large amount of images, which is almost impossible for users with lightweight devices (e.g. smart phones), so outsourcing image data to the cloud storage providers becomes one of the most convenient options because they provides enormous storage space and powerful computing ability. However, outsourcing the image data with sensitive information (such as financial position, personal identification and healthy records) to the server often results in great challenges in terms of data control and privacy. To prevent unauthorized access, the image owners usually encrypt their image data before its transmission to the server, but the conventional encryption operations pose a threat to image retrieval. For effectively utilizing and managing image resources, the researchers have made great efforts and they have proposed many practical retrieval schemes in this field [Reynolds (2016); Parikesit ], the image owners not only complete the task of index construction and encryption, but transmit the encrypted index to the server as well. Once receiving the query trapdoors, the server retrieves the intended images and returns them to the query users. These schemes provides feasible solutions and realize secure image retrieval in the encrypted domain, but the computation burden of index generation and encryption in the image client is really too heavy. To reduce the workload of image client and effectively implement retrieval work, some content-based image retrieval (CBIR) schemes ...

Adversarial attacks on fingerprint liveness detection

“…1. It is interesting that we notice the perturbation images show some similarity with the encrypted images [12][13][14][15][16], but the former are magnified noise while the latter are sophisticated designed encrypted files. Recent researchers have created serval methods to craft adversarial samples which vary greatly in terms of perturbation degree, number of perturbed pixels, and computation complexity.…”

Survey on JPEG compatible joint image compression and encryption algorithms

“…Additionally, joint compression and encryption methods in this scenario need to have efficient compression, so as to save the transmission bandwidth and storage space of data [88, 92, 93 ]. In [88, 92, 94, 103, 104 ], they proposed to realise cipher‐image retrieval based on encrypted DCT coefficients. This kind of algorithm relies on preserving the original distribution of the DCT coefficients, so permutation based operations in coefficients will be suitable.…”

Weather Recognition Based on 3C-CNN

“…Later, they simplified the classic GoogLeNet model and train it through a novel double-fine-tuning method. The model size is reduced to one third of the original GoogLeNet model, and the accuracy is increased from 94.5% to 95.46% [34]- [36]. In summary, although these methods based on deep learning are totally better than traditional methods, they can only run efficiently on GPU with high configuration, and need big data supporting, which makes weather identification costly.…”