Privacy-Preserving Extraction of HOG Features Based on Integer Vector Homomorphic Encryption

Yang, Haomiao; Huang, Yunfan; Yu, Yong; Yao, Mingxuan; Zhang, Xiaosong

doi:10.1007/978-3-319-72359-4_6

Cited by 16 publications

(11 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the last step, the client decrypts and recovers the feature descriptors combining both portions returned from the servers. Utilizing a more efficient homomorphic method (i.e., vector homomorphic encryption (VHE) [257]), Yang et al [240] also proposed a privacy-preserving scheme for extracting HOG features. The encryption is performed directly on the image vectors, which can be well applied for image processing.…”

Section: ) Image Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

A Comprehensive Survey on Secure Outsourced Computation and Its Applications

et al. 2019

View full text Add to dashboard Cite

With the ever-increasing requirement of storage and computation resources, it is unrealistic for local devices (with limited sources) to implement large-scale data processing. Therefore, individuals or corporations incline to outsource their computation requirements to the cloud. However, data outsourcing brings security and privacy concerns to users when the cloud servers are not fully trusted. Recently, extensive research works are conducted, aiming at secure outsourcing schemes for diverse computational tasks via different technologies. In this survey, we provide a technical review and comparison of existing outsourcing schemes using diverse secure computation methods. Specifically, we begin the survey by describing security threats and requirements of secure outsourcing computation. Meanwhile, we introduce four secure techniques (i.e., secure multi-party computation, pseudorandom functions, software guard extensions, and perturbation approaches) and their related works. Then, we focus on the theories and evolution of homomorphic encryption, as well as the applications of the basic operations and application-specific tasks. Finally, we discuss the security and performance of existing works and give future directions in this field. INDEX TERMS Secure outsourced computing, privacy preserving, homomorphic encryption, secure outsourced machine learning, data processing.

show abstract

Section: ) Image Feature Extractionmentioning

confidence: 99%

“…In the protocols, SURF and HOG tasks are implemented over rational and fixed-point binary numbers, respectively. Experimental evaluations demonstrated that the proposed solutions [239], [240], and [241] reach comparable performance to the original HOG solution.…”

Section: ) Image Feature Extractionmentioning

confidence: 99%

A Comprehensive Survey on Secure Outsourced Computation and Its Applications

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The HOG descriptor is one of the most popular approaches for object detection. It is invariant to illumination and geometric transformations and it has been successfully applied to many security applications, such as privacy in image feature extraction using homomorphic encryption [40], phishing detection [41], classification of sensitive information embedded within uploaded photos [42], handwritten digits recognition [43], facial expression recognition with CNNs [44] and, particularly, to face spoofing detection [11,[45][46][47][48]. Due to its popularity in anti-spoofing detection, in this work a variant of the HOG descriptor will be presented and experimentally validated.…”

Section: Related Workmentioning

confidence: 99%

Entropy-Based Face Recognition and Spoof Detection for Security Applications

2019

View full text Add to dashboard Cite

Nowadays, cyber attacks are becoming an extremely serious issue, which is particularly important to prevent in a smart city context. Among cyber attacks, spoofing is an action that is increasingly common in many areas, such as emails, geolocation services or social networks. Identity spoofing is defined as the action by which a person impersonates a third party to carry out a series of illegal activities such as committing fraud, cyberbullying, sextorsion, etc. In this work, a face recognition system is proposed, with an application to the spoofing prevention. The method is based on the Histogram of Oriented Gradients (HOG) descriptor. Since different face regions do not have the same information for the recognition process, introducing entropy would quantify the importance of each face region in the descriptor. Therefore, entropy is added to increase the robustness of the algorithm. Regarding face recognition, our approach has been tested on three well-known databases (ORL, FERET and LFW) and the experiments show that adding entropy information improves the recognition rate significantly, with an increase over 40% in some of the considered databases. Spoofing tests has been implemented on CASIA FASD and MIFS databases, having obtained again better results than similar texture descriptors approaches.

show abstract

“…The methods are classified into two types: perceptual encryption-based type [2][3][4][5] and homomorphic encryption (HE)-based one. [6][7][8][9] In recent years, considerable efforts have been made in the fields of fully HE and multi-party computation. 10 Some attempts with HE-based one have been made on learning.…”

Section: Introductionmentioning

confidence: 99%

“…10 Some attempts with HE-based one have been made on learning. [6][7][8][9] However, HE-based schemes require algorithms specialized for computing encrypted data be prepared, and high computational complexity. 7 In addition, it is also difficult to maintain high computational accuracy.…”

Section: Introductionmentioning

confidence: 99%

Privacy-preserving machine learning using EtC images

Kawamura

Kinoshita

Kiya

2020

International Workshop on Advanced Imaging Technology (IWAIT) 2020

View full text Add to dashboard Cite

We propose a privacy-preserving machine learning scheme with encryption-then-compression (EtC) images, where EtC images are images encrypted by using a block-based encryption method proposed for EtC systems with JPEG compression. In this paper, a novel property of EtC images is first discussed, although EtC ones was already shown to be compressible as a property. The novel property allows us to directly apply EtC images to machine learning algorithms non-specialized for computing encrypted data. In addition, the proposed scheme is demonstrated to provide no degradation in the performance of some typical machine learning algorithms including the support vector machine algorithm with kernel trick and random forests under the use of z-score normalization. A number of facial recognition experiments with are carried out to confirm the effectiveness of the proposed scheme.

show abstract

Privacy-Preserving Extraction of HOG Features Based on Integer Vector Homomorphic Encryption

Cited by 16 publications

References 16 publications

A Comprehensive Survey on Secure Outsourced Computation and Its Applications

A Comprehensive Survey on Secure Outsourced Computation and Its Applications

Entropy-Based Face Recognition and Spoof Detection for Security Applications

Privacy-preserving machine learning using EtC images

Contact Info

Product

Resources

About