{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T04:19:45Z","timestamp":1741753185944,"version":"3.38.0"},"reference-count":49,"publisher":"SAGE Publications","issue":"4","license":[{"start":{"date-parts":[[2022,11,21]],"date-time":"2022-11-21T00:00:00Z","timestamp":1668988800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Journal of Computer Security"],"published-print":{"date-parts":[[2023,8,14]]},"abstract":"<jats:p> Due to the limited capabilities of user devices, such as smart phones, and the Internet of Things (IoT), edge intelligence is being recognized as a promising paradigm to enable effective analysis of the data generated by these devices with complex artificial intelligence (AI) models, and it often entails either fully or partially offloading the computation of neural networks from user devices to edge computing servers. To protect users\u2019 data privacy in the process, most existing researches assume that the private (sensitive) attributes of user data are known in advance when designing privacy-protection measures. This assumption is restrictive in real life, and thus limits the application of these methods. Inspired by the research in image steganography and cyber deception, in this paper, we propose StegEdge, a conceptually novel approach to this challenge. StegEdge takes as input the user-generated image and a randomly selected \u201ccover\u201d image that does not pose any privacy concern (e.g., downloaded from the Internet), and extracts the features such that the utility tasks can still be conducted by the edge computing servers, while potential adversaries seeking to reconstruct\/recover the original user data or analyze sensitive attributes from the extracted features sent from users to the server, will largely acquire information of the cover image. Thus, users\u2019 data privacy is protected via a form of deception. Empirical results conducted on the CelebA and ImageNet datasets show that, at the same level of accuracy for utility tasks, StegEdge reduces the adversaries\u2019 accuracy of predicting sensitive attributes by up to 38% compared with other methods, while also defending against adversaries seeking to reconstruct user data from the extracted features. <\/jats:p>","DOI":"10.3233\/jcs-220042","type":"journal-article","created":{"date-parts":[[2022,11,22]],"date-time":"2022-11-22T16:08:37Z","timestamp":1669133317000},"page":"393-417","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":1,"title":["StegEdge: Privacy protection of unknown sensitive attributes in edge intelligence via deception"],"prefix":"10.1177","volume":"31","author":[{"given":"Jianfeng","family":"Zhang","sequence":"first","affiliation":[{"name":"College of Computer Science, Nankai University, Tianjin, China"}]},{"given":"Wensheng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Automation, Chinese Academy of Sciences, Beijing, China"}]},{"given":"Jingdong","family":"Xu","sequence":"additional","affiliation":[{"name":"College of Computer Science, Nankai University, Tianjin, China"}]}],"member":"179","published-online":{"date-parts":[[2022,11,21]]},"reference":[{"key":"ref001","doi-asserted-by":"crossref","unstructured":"A.Achille and S.Soatto, Information dropout: Learning optimal representations through noisy computation, 40(12) (2018), 2897\u20132905.","DOI":"10.1109\/TPAMI.2017.2784440"},{"key":"ref002","doi-asserted-by":"crossref","unstructured":"M.Andriushchenko, F.Croce, N.Flammarion and M.Hein, Square attack: A query-efficient black-box adversarial attack via random search, in: European Conference on Computer Vision, Springer, 2020, pp. 484\u2013501.","DOI":"10.1007\/978-3-030-58592-1_29"},{"issue":"5","key":"ref003","first-page":"432","volume":"12","author":"Beaudry N.J.","year":"2012","journal-title":"Quantum Info. Comput."},{"key":"ref004","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0177678"},{"key":"ref005","doi-asserted-by":"crossref","unstructured":"J.W.Chen, L.J.Chen, C.M.Yu and C.S.Lu, Perceptual Indistinguishability-Net (PI-Net): Facial image obfuscation with manipulable semantics, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 6478\u20136487.","DOI":"10.1109\/CVPR46437.2021.00641"},{"key":"ref006","doi-asserted-by":"crossref","unstructured":"L.Chen, S.Lin, X.Lu, D.Cao, H.Wu, C.Guo, C.Liu and F.Y.Wang, Deep neural network based vehicle and pedestrian detection for autonomous driving: A survey, 22(6) (2021), 3234\u20133246.","DOI":"10.1109\/TITS.2020.2993926"},{"key":"ref007","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref008","doi-asserted-by":"crossref","unstructured":"S.Deng, H.Zhao, W.Fang, J.Yin, S.Dustdar and A.Y.Zomaya, Edge intelligence: The confluence of edge computing and artificial intelligence, 7(8) (2020), 7457\u20137469.","DOI":"10.1109\/JIOT.2020.2984887"},{"key":"ref009","doi-asserted-by":"crossref","unstructured":"Y.Du, J.Xu, H.Xiong, Q.Qiu, X.Zhen, C.G.Snoek and L.Shao, Learning to learn with variational information bottleneck for domain generalization, in: European Conference on Computer Vision, Springer, 2020, pp. 200\u2013216.","DOI":"10.1007\/978-3-030-58607-2_12"},{"key":"ref010","doi-asserted-by":"crossref","unstructured":"X.Duan, D.Guo, N.Liu, B.Li, M.Gou and C.Qin, A new high capacity image steganography method combined with image elliptic curve cryptography and deep neural network, 8 (2020), 25777\u201325788.","DOI":"10.1109\/ACCESS.2020.2971528"},{"key":"ref011","doi-asserted-by":"crossref","unstructured":"A.Elad, D.Haviv, Y.Blau and T.Michaeli, Direct validation of the information bottleneck principle for deep nets, in: Proceedings of the IEEE\/CVF International Conference on Computer Vision Workshops, 2019.","DOI":"10.1109\/ICCVW.2019.00099"},{"key":"ref012","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2019.2947893"},{"key":"ref013","unstructured":"K.J.Ferguson-Walter, M.M.Major, C.K.Johnson and D.H.Muhleman, Examining the efficacy of decoy-based and psychological cyber deception, in: 30th USENIX Security Symposium (USENIX Security, Vol. 21, 2021, pp. 1127\u20131144."},{"key":"ref014","doi-asserted-by":"crossref","unstructured":"H.Guo, B.Dolhansky, E.Hsin, P.Dinh, C.C.Ferrer and S.Wang, Deep poisoning: Towards robust image data sharing against visual disclosure, in: Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision, 2021, pp. 686\u2013696.","DOI":"10.1109\/WACV48630.2021.00073"},{"key":"ref015","doi-asserted-by":"crossref","unstructured":"Z.He, T.Zhang and R.B.Lee, Attacking and protecting data privacy in edge\u2013cloud collaborative inference systems, 8(12) (2020), 9706\u20139716.","DOI":"10.1109\/JIOT.2020.3022358"},{"key":"ref016","doi-asserted-by":"crossref","unstructured":"A.Howard, M.Sandler, G.Chu, L.C.Chen, B.Chen, M.Tan, W.Wang, Y.Zhu, R.Pang, V.Vasudevan et al., Searching for mobilenetv3, in: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2019, pp. 1314\u20131324.","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref017","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737614"},{"key":"ref018","unstructured":"IDC, European Enterprise Edge Market Forecast 2020\u20132024 (2021), https:\/\/www.idc.com\/getdoc.jsp?containerId=EUR147186321."},{"key":"ref019","doi-asserted-by":"publisher","DOI":"10.1109\/SecDev45635.2020.00023"},{"key":"ref020","doi-asserted-by":"crossref","unstructured":"Q.Jin, L.Yang and Z.Liao, Adabits: Neural network quantization with adaptive bit-widths, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 2146\u20132156.","DOI":"10.1109\/CVPR42600.2020.00222"},{"key":"ref021","doi-asserted-by":"crossref","unstructured":"J.Johnson, A.Alahi and L.Fei-Fei, Perceptual losses for real-time style transfer and super-resolution, in: European Conference on Computer Vision, Springer, 2016, pp. 694\u2013711.","DOI":"10.1007\/978-3-319-46475-6_43"},{"key":"ref022","unstructured":"D.Jung, H.Bae, H.S.Choi and S.Yoon, PixelSteganalysis: Pixel-wise hidden information removal with low visual degradation, 2021."},{"key":"ref023","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2018.06.075"},{"key":"ref024","doi-asserted-by":"crossref","unstructured":"A.Li, Y.Duan, H.Yang, Y.Chen and J.Yang, TIPRDC: Task-independent privacy-respecting data crowdsourcing framework for deep learning with anonymized intermediate representations, in: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2020, pp. 824\u2013832.","DOI":"10.1145\/3394486.3403125"},{"key":"ref025","doi-asserted-by":"crossref","unstructured":"Y.Li, X.Xu, J.Xiao, S.Li and H.T.Shen, Adaptive square attack: Fooling autonomous cars with adversarial traffic signs, 8(8) (2020), 6337\u20136347.","DOI":"10.1109\/JIOT.2020.3016145"},{"key":"ref026","doi-asserted-by":"crossref","unstructured":"W.Y.B.Lim, J.S.Ng, Z.Xiong, J.Jin, Y.Zhang, D.Niyato, C.Leung and C.Miao, Decentralized edge intelligence: A dynamic resource allocation framework for hierarchical federated learning, 33(3) (2021), 536\u2013550.","DOI":"10.1109\/TPDS.2021.3096076"},{"key":"ref027","doi-asserted-by":"crossref","unstructured":"Z.Liu, P.Luo, X.Wang and X.Tang, Deep learning face attributes in the wild, in: Proceedings of International Conference on Computer Vision (ICCV), 2015.","DOI":"10.1109\/ICCV.2015.425"},{"key":"ref028","doi-asserted-by":"crossref","unstructured":"L.McInnes, J.Healy and J.Melville, UMAP: Uniform manifold approximation and projection for dimension reduction, 2018, arXiv preprint arXiv:1802.03426.","DOI":"10.21105\/joss.00861"},{"key":"ref029","doi-asserted-by":"crossref","unstructured":"V.Mirjalili, S.Raschka and A.Ross, PrivacyNet: Semi-adversarial networks for multi-attribute face privacy, 29 (2020), 9400\u20139412.","DOI":"10.1109\/TIP.2020.3024026"},{"key":"ref030","doi-asserted-by":"crossref","unstructured":"P.Mishra, R.Lehmkuhl, A.Srinivasan, W.Zheng and R.A.Popa, Delphi: A cryptographic inference service for neural networks, in: 29th USENIX Security Symposium (USENIX Security, Vol. 20, 2020, pp. 2505\u20132522.","DOI":"10.1145\/3411501.3419418"},{"key":"ref031","doi-asserted-by":"crossref","unstructured":"A.Morales, J.Fierrez, R.Vera-Rodriguez and R.Tolosana, Sensitivenets: Learning agnostic representations with application to face images, 43(6) (2020), 2158\u20132164.","DOI":"10.1109\/TPAMI.2020.3015420"},{"key":"ref032","doi-asserted-by":"publisher","DOI":"10.1145\/3439950"},{"key":"ref033","doi-asserted-by":"crossref","unstructured":"W.Qi, Q.Li, Q.Song, L.Guo and A.Jamalipour, Extensive edge intelligence for future vehicular networks in 6G, 28(4) (2021), 128\u2013135.","DOI":"10.1109\/MWC.001.2000393"},{"key":"ref034","doi-asserted-by":"publisher","DOI":"10.1145\/3372297.3417274"},{"key":"ref035","unstructured":"M.Samragh, H.Hosseini, A.Triastcyn, K.Azarian, J.Soriaga and F.Koushanfar, Unsupervised information obfuscation for split inference of neural networks, in: International Conference on Machine Learning, PMLR, 2021."},{"key":"ref036","doi-asserted-by":"crossref","unstructured":"S.Sannon, N.N.Bazarova and D.Cosley, Privacy lies: Understanding how, when, and why people Lie to protect their privacy in multiple online contexts, in: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, 2018, pp. 1\u201313.","DOI":"10.1145\/3173574.3173626"},{"key":"ref037","doi-asserted-by":"publisher","DOI":"10.1145\/3411508.3421376"},{"key":"ref038","unstructured":"T.Speicher, M.Ali, G.Venkatadri, F.N.Ribeiro, G.Arvanitakis, F.Benevenuto, K.P.Gummadi, P.Loiseau and A.Mislove, Potential for discrimination in online targeted advertising, in: Conference on Fairness, Accountability and Transparency, PMLR, 2018, pp. 5\u201319."},{"key":"ref039","unstructured":"T.Standley, A.Zamir, D.Chen, L.Guibas, J.Malik and S.Savarese, Which tasks should be learned together in multi-task learning? in: International Conference on Machine Learning, PMLR, 2020, pp. 9120\u20139132."},{"key":"ref040","doi-asserted-by":"crossref","unstructured":"X.Sun, G.Cheng, H.Li, L.Pei and J.Han, Exploring effective data for surrogate training towards black-box attack, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 15355\u201315364.","DOI":"10.1109\/CVPR52688.2022.01492"},{"key":"ref041","unstructured":"F.Tram\u00e8r and D.Boneh, Slalom: Fast, verifiable and private execution of neural networks in trusted hardware, in: 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6\u20139, 2019, OpenReview.net, 2019."},{"issue":"11","key":"ref042","first-page":"2579","volume":"9","author":"Van der Maaten L.","year":"2008","journal-title":"Journal of machine learning research"},{"key":"ref043","unstructured":"Z.Wu, H.Wang, Z.Wang, H.Jin and Z.Wang, Privacy-preserving deep action recognition: An adversarial learning framework and a new dataset, 2020."},{"key":"ref044","doi-asserted-by":"crossref","unstructured":"T.Xiao, Y.H.Tsai, K.Sohn, M.Chandraker and M.H.Yang, Adversarial learning of privacy-preserving and task-oriented representations, in: Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 34, 2020, pp. 12434\u201312441.","DOI":"10.1609\/aaai.v34i07.6930"},{"key":"ref045","doi-asserted-by":"crossref","unstructured":"Y.Xiao, G.Shi, Y.Li, W.Saad and H.V.Poor, in: Toward Self-Learning Edge Intelligence in 6G 58(12), 2020, pp. 34\u201340.","DOI":"10.1109\/MCOM.001.2000388"},{"key":"ref046","doi-asserted-by":"crossref","unstructured":"G.Yang, N.Fei, M.Ding, G.Liu, Z.Lu and T.Xiang, L2m-gan: Learning to manipulate latent space semantics for facial attribute editing, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 2951\u20132960.","DOI":"10.1109\/CVPR46437.2021.00297"},{"key":"ref047","doi-asserted-by":"crossref","unstructured":"P.Yang, Y.Lao and P.Li, Robust watermarking for deep neural networks via bi-level optimization, in: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2021, pp. 14841\u201314850.","DOI":"10.1109\/ICCV48922.2021.01457"},{"key":"ref048","doi-asserted-by":"crossref","unstructured":"X.Zhou, X.Xu, W.Liang, Z.Zeng and Z.Yan, Deep-learning-enhanced multitarget detection for end\u2013edge\u2013cloud surveillance in smart IoT, 8(16) (2021), 12588\u201312596.","DOI":"10.1109\/JIOT.2021.3077449"},{"key":"ref049","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2019.2918951"}],"container-title":["Journal of Computer Security"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JCS-220042","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.3233\/JCS-220042","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JCS-220042","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,11]],"date-time":"2025-03-11T06:14:55Z","timestamp":1741673695000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.3233\/JCS-220042"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,21]]},"references-count":49,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2023,8,14]]}},"alternative-id":["10.3233\/JCS-220042"],"URL":"https:\/\/doi.org\/10.3233\/jcs-220042","relation":{},"ISSN":["0926-227X","1875-8924"],"issn-type":[{"type":"print","value":"0926-227X"},{"type":"electronic","value":"1875-8924"}],"subject":[],"published":{"date-parts":[[2022,11,21]]}}}