Paper Infomation
Research on Privacy Protection of Intelligent Applications
Full Text(PDF, 2983KB)
Author: Shengdi Zhao, Yanxin Yao
Abstract: There are many privacy protection methods in the field of artificial intelligence. Firstly, this paper summarizes the related secure multi-party privacy computing methods, image retrieval privacy protection methods, and machine learning privacy protection methods. At present, edge computing provides many benefits for various intelligent applications, but at the same time, when end-to-edge distributed computing is carried out during the unloading process of edge computing, privacy disclosure will occur. In this paper, a distributed layout privacy protection strategy is proposed to ensure the two-way tasks of face attribute feature extraction and privacy feature hiding. The main purpose is to avoid the remote transmission of privacy information characters while transmitting the main tasks, and to eliminate the hidden processing on mobile devices as much as possible, so as to improve the effectiveness of privacy protection. From the final experimental results, it can be concluded that the network framework algorithm can effectively achieve the effect of privacy blanking.
Keywords: Edge Computing, Privacy Concealment, Gradient Flipping Layer, Local Fine-tuning
References:
[1] H Li, G Shou, Y Hu, et al. Mobile edge computing: Progress and challenges[C]. 2016 4th IEEE International Conference on Mobile Cloud Computing, Services, and Engineering (MobileCloud), 2016: 83-84.
[2] Ma L, Q Pei, H Xiao, et al. Edge computing enhanced privacy preserving for location-based services[C]. IEEE INFOCOM 2019-IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), 2019: 1-6.
[3] X Xu, C He, Z Xu, et al. Joint optimization of offloading utility and privacy for edge computing enabled IoT[J]. IEEE Internet of Things Journal, 2019, 7(4): 2622-2629.
[4] M Sharif, S Bhagavatula, L Bauer, et al. Accessorize to a crime: Real and stealthy attacks on state-of-the-art face recognition[C]. Proceedings of the 2016 Acm Sigsac Conference on Computer and Communications Security, 2016: 1528-1540.
[5] D J Robertson, A M Burton. Unfamiliar face recognition: Security, surveillance and smartphones[J]. The Journal of the Homeland Defense and Security Information Analysis Center, 2016: 14-21.
[6] S C Sen-ching, M U Rafique, W Tan. Privacy-preserving distributed deep learning with privacy transformations[C]. 2018 IEEE International Workshop on Information Forensics and Security (WIFS), 2018: 1-7.
[7] Z Chen, L Li, H Peng, et al. An evaluation method of image scrambling degree based on pixel distribution[C]. 2018 International Symposium on Security and Privacy in Social Networks and Big Data (SocialSec), 2018: 206-211.
[8] W Zhang, H Wang, D Hou, et al. Reversible data hiding in encrypted images by reversible image transformation[J]. IEEE Transactions on Multimedia, 2016, 18(8): 1469-1479.
[9] Y Ganin, V Lempitsky. Unsupervised domain adaptation by backpropagation[C]. International Conference on Machine Learning, PMLR, 2015: 1180-1189.
[10] L Cui, Z Chen, A Wang, et al. Development of a robust cooperative adaptive cruise control with dynamic topology[J]. IEEE Transactions on Intelligent Transportation Systems, 2021: 1-12.
[11] Z Chen, B B Park. Cooperative adaptive cruise control with unconnected vehicle in the loop[J]. IEEE Transactions on Intelligent Transportation Systems, 2020: 1-11.
[12] K Karthik, S Kashyap. Transparent hashing in the encrypted domain for privacy preserving image retrieval[J]. Signal, Image and Video Processing, 2013, 7(4): 647-664.
[13] W Lu, A L Varna, M Wu. Confidentiality-preserving image search: A comparative study between homomorphic encryption and distance-preserving randomization[J]. IEEE Access, 2014, 2: 125-141.
[14] H Akbari-Nodehi, M A Maddah-Ali. Secure coded multi-party computation for massive matrix operations[J]. IEEE Transactions on Information Theory, 2021, 67(4): 2379-2398.
[15] P Ah-Fat, M Huth. Optimal accuracy-privacy trade-off for secure computations[J]. IEEE Transactions on Information Theory, 2018, 65(5): 3165-3182.
[16] 杨乾. 一种新的人脸识别隐私保护方案[D]. 华中师范大学, 2020.
[17] Y Cai, C Tang. Securely outsourced face recognition under federated cloud environment[C]. 2016 15th International Symposium on Parallel and Distributed Computing (ISPDC), 2016: 269-276.
[18] S Pearson, A Charlesworth. Accountability as a way forward for privacy protection in the cloud[C]. IEEE International Conference on Cloud Computing, Springer, Berlin, Heidelberg, 2009: 131-144.
[19] Y Ma, L Wu, X Gu, et al. A secure face-verification scheme based on homomorphic encryption and deep neural networks[J]. IEEE Access, 2017, 5: 16532-16538.
[20] M Haghighat, S Zonouz, M Abdel-Mottaleb. CloudID: Trustworthy cloud-based and cross-enterprise biometric identification[J]. Expert Systems with Applications, 2015, 42(21): 7905-7916.
[21] 彭远帆. 隐私保护的图像检索关键技术研究[D]. 北京工业大学, 2015.
[22] H Akbari-Nodehi, M A Maddah-Ali. Secure coded multi-party computation for massive matrix operations[J]. IEEE Transactions on Information Theory, 2021, 67(4): 2379-2398.
[23] Z Huang, M Zhang, Y Zhang. Toward efficient encrypted image retrieval in cloud environment[J]. IEEE Access, 2019, 7: 174541-174550.
[24] Z Xia, L Jiang, D Liu, et al. BOEW: A content-based image retrieval scheme using bag-of-encrypted-words in cloud computing[J]. IEEE Transactions on Services Computing, 2022, 15(1): 202-214.
[25] B Ferreira, J Rodrigues, J Leitão, et al. Towards an image encryption scheme with content-based image retrieval properties[M]. Data Privacy Management, Autonomous Spontaneous Security, and Security Assurance, Springer, Cham, 2014: 311-318.
[26] Z Xia, N N Xiong, A V Vasilakos, et al. EPCBIR: An efficient and privacy-preserving content-based image retrieval scheme in cloud computing[J]. Information Sciences, 2017, 387: 195-204.
[27] Y Xu, J Gong, L Xiong, et al. A privacy-preserving content-based image retrieval method in cloud environment[J]. Journal of Visual Communication and Image Representation, 2017, 43: 164-172.
[28] 唐鹏, 黄征, 邱卫东. 深度学习中的隐私保护技术综述[J]. 信息安全与通信保密, 2019, (6): 55-62.
[29] S Abd Razak, N H M Nazari, A Al-Dhaqm. Data anonymization using pseudonym system to preserve data privacy[J]. IEEE Access, 2020, 8: 43256-43264.
[30] Y Aono, T Hayashi, L Wang, et al. Privacy-preserving deep learning via additively homomorphic encryption[J]. IEEE Transactions on Information Forensics and Security, 2017, 13(5): 1333-1345.
[31] G S Uehara, A Spanias, W Clark. Quantum information processing algorithms with emphasis on machine learning[C]. 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA), 2021: 1-11.
[32] 李晓东, 韩青, 金鑫. 一种基于深度学习和同态加密的安全高效的人脸识别方法[P]: CN, CN201810973325.6, 2019-1-4.
[33] N Papernot, S Song, I Mironov, et al. Scalable private learning with pate[J]. 2018: arXiv: 1802.08908.
[34] H Wang, Y Zhang, L You, et al. Image privacy perception method based on deep learning[P]: US, US16099836. 2021-7-22.