Catch me in the dark: Effective privacy-preserving outsourcing of feature extractions over image data

Wang, Qian; Hu, Shengshan; Ren, Kui; Wang, Jingjun; Wang, Zhibo; Du, Minxin

doi:10.1109/infocom.2016.7524460

Cited by 31 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, researchers have extended privacy-preserving information retrieval from search over encrypted text [3-5, 7, 23] to secure multimedia retrieval [8,[24][25][26][27][28][29][30][31].…”

Section: Related Workmentioning

confidence: 99%

Secure Outsourcing Algorithm of BTC Feature Extraction in Cloud Computing

Jiang

Yang

2020

IEEE Access

View full text Add to dashboard Cite

Advances in cloud computing have aroused many researchers' interest in privacy-preserving feature extraction over outsourced multimedia data, especially private image data. Since block truncation coding (BTC) is known as a simple and efficient technology for image compression, this paper focuses on privacy-preserving feature extraction in BTC compressed domain. We propose a privacy-preserving computation of BTC feature extraction over massive encrypted images (also called PPBTC). First, all images are uploaded to the cloud after encryption. The privacy-preserving image encryption process consists of block permutation, pixel diffusion, and a bit-plane random shift. BTC features remain unchanged after encryption and the cloud server can directly extract BTC features from the encrypted images. Some analyses and experimental results demonstrate that the proposed privacy-preserving feature extraction scheme for BTCcompressed images is efficient and secure, and it can be applied to secure image computation applications in cloud computing. INDEX TERMS Feature extraction, privacy-preserving, block truncation coding, homomorphic encryption.

show abstract

Section: Related Workmentioning

confidence: 99%

Secure Outsourcing Algorithm of BTC Feature Extraction in Cloud Computing

Jiang

Yang

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…A number of studies proposed cryptography-based solutions for image sharing [7,8], retrieval [9,10], and feature extraction [11,12] using untrusted service providers. While those solutions secure the image data with encryption, they exhibit a few drawbacks which make them inapplicable in our setting.…”

Section: Introductionmentioning

confidence: 99%

“…Secondly, the features computed by the untrusted server also need to be protected, such as shape positions and scale-invariant feature transform (SIFT), as those features often disclose sensitive information. Existing studies resort to more expensive cryptographic tools, such as homomorphic encryption and garbled circuit [11], or multiple independent servers [12], which potentially limit the feasibility of extracting complex features and enabling time-critical applications.…”

Section: Introductionmentioning

confidence: 99%

Image Pixelization with Differential Privacy

Fan

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Ubiquitous surveillance cameras and personal devices have given rise to the vast generation of image data. While sharing the image data can benefit various applications, including intelligent transportation systems and social science research, those images may capture sensitive individual information, such as license plates, identities, etc. Existing image privacy preservation techniques adopt deterministic obfuscation, e.g., pixelization, which can lead to re-identification with well-trained neural networks. In this study, we propose sharing pixelized images with rigorous privacy guarantees. We extend the standard differential privacy notion to image data, which protects individuals, objects, or their features. Empirical evaluation with real-world datasets demonstrates the utility and efficiency of our method; despite its simplicity, our method is shown to effectively reduce the success rate of re-identification attacks.

show abstract

“…It should be noticed that both solutions given in [12] and [13] do not well preserve the performance of the original SIFT features. To overcome this issue, [15] and [16] propose a secure SIFT feature set based on two independent cloud servers. If the complexity of communication with the user is reduced, it assumes that the two servers do not collude.…”

Section: Introductionmentioning

confidence: 99%

“…There is thus a need to extract image signatures that are themselves encrypted. Such an issue was very recently addressed in [21] and by next in [15] [16]. But, these solutions exploit at least two cloud servers when computing signatures with as consequence the increase of already high communication requirements.…”

Section: Introductionmentioning

confidence: 99%

Secured Outsourced Content Based Image Retrieval Based on Encrypted Signatures Extracted From Homomorphically Encrypted Images

Bellafqira,

Coatrieux,

Bouslimi

et al. 2017

Preprint

View full text Add to dashboard Cite

In this paper, we present a novel Secured Outsourced Content Based Image Retrieval solution, which allows looking for similar images stored into the cloud in a homomorphically encrypted form. Its originality is fourfold. In a first time, it extracts from a Paillier encrypted image a wavelet based global image signature. In second, this signature is extracted in an encrypted form and gives no clues about the image content. In a third time, its calculation does not require the cloud to communicate with a trusted third party as usually proposed by other existing schemes. More clearly, all computations required in order to look for similar images are conducted by the cloud only with no extra-communications. To make possible such a computation, we propose a new fast way to compare encrypted data, these ones being encrypted by the same public key or not, and using a recursive relationship in-between Paillier random values when computing the different resolution levels of the image wavelet transform. Experiments conducted in two distinct frameworks: medical image retrieval with as purpose diagnosis aid support, and face recognition for user authentication; indicate that the proposed SOCBIR does not change image retrieval performance.

show abstract

Catch me in the dark: Effective privacy-preserving outsourcing of feature extractions over image data

Cited by 31 publications

References 13 publications

Secure Outsourcing Algorithm of BTC Feature Extraction in Cloud Computing

Secure Outsourcing Algorithm of BTC Feature Extraction in Cloud Computing

Image Pixelization with Differential Privacy

Secured Outsourced Content Based Image Retrieval Based on Encrypted Signatures Extracted From Homomorphically Encrypted Images

Contact Info

Product

Resources

About