Page 108 - 《软件学报》2026年第6期

肖欣怡 等: 恶意敌手环境下的隐私保护目标检测 2427

Enhancing Technologies, 2020, 2020(2): 459–480. [doi: 10.2478/popets-2020-0036]
[27] Wagh S, Tople S, Benhamouda F, Kushilevitz E, Mittal P, Rabin T. Falcon: Honest-majority maliciously secure framework for private
deep learning. Proc. on Privacy Enhancing Technologies, 2021, 2021(1): 188–208. [doi: 10.2478/popets-2021-0011]
[28] Dalskov A, Escudero D, Keller M. Fantastic four: Honest-majority four-party secure computation with malicious security. In: Proc. of the
30th USENIX Security Symp. Vancouver: USENIX Association, 2021. 2183–2200.
[29] Keller M. MP-SPDZ: A versatile framework for multi-party computation. In: Proc. of the 2020 ACM SIGSAC Conf. on Computer and
Communications Security. ACM, 2020. 1575–1590. [doi: 10.1145/3372297.3417872]
[30] Girshick R. Fast R-CNN. In: Proc. of the 2015 IEEE Int’l Conf. on Computer Vision. Santiago: IEEE, 2015. 1440–1448. [doi: 10.1109/
ICCV.2015.169]
[31] Ren SQ, He KM, Girshick R, Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Trans. on
Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137–1149. [doi: 10.1109/tpami.2016.2577031]
[32] Lin TY, Dollár P, Girshick R, He KM, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: Proc. of the 2017
IEEE Conf. on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017. 936–944. [doi: 10.1109/CVPR.2017.106]
[33] Chen MQ, Yu LJ, Zhi C, Sun RJ, Zhu SW, Gao ZY, Ke ZX, Zhu MQ, Zhang YM. Improved Faster R-CNN for fabric defect detection
based on Gabor filter with genetic algorithm optimization. Computers in Industry, 2022, 134: 103551. [doi: 10.1016/j.compind.2021.
103551]
[34] Keller M, Sun K. Secure quantized training for deep learning. In: Proc. of the 39th Int’l Conf. on Machine Learning. Baltimore: PMLR,
2022. 10912–10938.
[35] Faraji S, Kerschbaum F. Trifecta: Faster high-throughput three-party computation over WAN using multi-fan-in logic gates. Proc. on
Privacy Enhancing Technologies, 2023, 2023(4): 224–237. [doi: 10.56553/popets-2023-0107]
[36] Liu FR, Xie X, Yu Y. Scalable multi-party computation protocols for machine learning in the honest-majority setting. In: Proc. of the
33rd USENIX Security Symp. Philadelphia: USENIX Association, 2024. 109.
[37] Aly A, Smart NP. Benchmarking privacy preserving scientific operations. In: Proc. of the 17th Int’l Conf. on Applied Cryptography and
Network Security. Bogota: Springer, 2019. 509–529. [doi: 10.1007/978-3-030-21568-2_25]
[38] Koti N, Pancholi M, Patra A, Suresh A. SWIFT: Super-fast and robust privacy-preserving machine learning. In: Proc. of the 30th
USENIX Security Symp. Vancouver: USENIX Association, 2021. 2651–2668.
[39] Ma Z, Liu Y, Liu XM, Ma JF, Li FF. Privacy-preserving outsourced speech recognition for smart IoT devices. IEEE Internet of Things
Journal, 2019, 6(5): 8406–8420. [doi: 10.1109/JIOT.2019.2917933]
[40] Liu L, Fu SJ, Huang XL, Luo YC, Zhang XY, Choo KKR. SecDM: A secure and lossless human mobility prediction system. IEEE Trans.
on Services Computing, 2024, 17(4): 1793–1805. [doi: 10.1109/TSC.2024.3358292]

附中文参考文献
[9] 廖育荣, 王海宁, 林存宝, 李阳, 方宇强, 倪淑燕. 基于深度学习的光学遥感图像目标检测研究进展. 通信学报, 2022, 43(5): 190–203.
[doi: 10.11959/j.issn.1000−436x.2022071]
[18] 白利芳, 祝跃飞, 李勇军, 王帅, 杨晓琪. 全同态加密研究进展. 计算机研究与发展, 2024, 61(12): 3069–3087. [doi: 10.7544/issn1000-
1239.202221052]
[20] 李梦倩, 田有亮, 张军鹏, 赵冬梅. 基于零集中差分隐私的联邦学习激励方案. 通信学报, 2025, 46(1): 79–92. [doi: 10.11959/j.issn.1000-
436x.2025008]
[21] 张朋飞, 朱伊波, 程祥, 张治坤, 刘西蒙, 孙笠, 方贤进, 张吉. 基于自适应剪枝的满足本地差分隐私的真值发现算法. 软件学报, 2025,
36(7): 3405–3428. http://www.jos.org.cn/1000-9825/7287.htm [doi: 10.13328/j.cnki.jos.007287]
[25] 刘伟欣, 管晔玮, 霍嘉荣, 丁元朝, 郭华, 李博. 一种基于安全多方计算的快速 Transformer 安全推理方案. 计算机研究与发展, 2024,
61(5): 1218–1229. [doi: 10.7544/issn1000-1239.202330966]

附录 A 正确性与安全性分析

本附录将对 MalOD 的正确性和安全性进行证明.
A.1 MalOD 的正确性
在 MalOD 中, 每个预处理后的图像被分为 4 个加法份额: I 、 I 、 I (2) 和 I , 然后依次通过 SecBackbone、
(3)
(1)
(0)