基于注意力模块的移动设备多场景持续身份认证

doi:10.3778/j.issn.1002-8331.2209-0204

摘要/Abstract

摘要： 针对用户与移动设备交互时会产生场景变化，现有工作中只能采集特定的单一场景特征，无法实现多场景转换认证，并且身份认证准确率较低的问题，提出了一种基于移动模式的、注意力模块和卷积神经网络融合（CNN-SACA）的多场景持续认证方案。在不限使用场景和操作的情况下，提取用户与移动设备交互时的移动模式（movement patterns，MP）特征，捕捉在动态和静态场景下产生的手部微运动，从而实现多场景的身份认证。设计并使用了一个包括五层卷积层结构的卷积神经网络，在第一层卷积后按序通过改进的空间和通道注意力子模块，再在多层卷积后进行反序分配权重，从两个维度来对通过卷积后所表征的MP特征分配双重注意力权重，增强关键特征表达。利用公开数据集验证所提方案在多场景身份认证方面的有效性和可行性。实验结果表明，所提出的基于移动模式的深度学习模型可以较好地解决身份认证场景单一的局限性，多场景的身份认证的准确率达到99.6%；同时，所提出的CNN-SACA模型与单独的CNN模型相比准确率提高了1.5个百分点，有效改善多场景下的移动设备身份认证能力。

关键词: 卷积神经网络, 注意力模块, 多场景, 持续身份认证, 移动设备

Abstract: In view of the fact that the user may change the scene when interacting with the mobile device, the existing works have limitations on the specific single scene when collecting features and low authentication accuracies, and cannot achieve multi-scene conversion authentication. To overcome these issues，a movement patterns based multi-scene continuous authentication scheme, which combines the attention module with the convolutional neural network (CNN-SACA) is proposed. Under unrestricted usage scenarios and operations, the movement patterns (MP) features are extracted when the user interacts with the mobile device and then hand micro-motion can be captured in dynamic and static scenes, by which the multi-scene authentication is realized. A convolutional neural network including 5 convolutional layers is designed. After the convolution of the first layer, the improved spatial and channel attention sub modules are sequentially passed, and then the weights are inversely distributed after the convolution of the multiple layers to enhance the key feature representation. MP features characterized by the convolution are assigned double attention weights from two dimensions. A public data set is used to verify the effectiveness and feasibility of the proposed method in multi-scene authentication. The experimental results show that the proposed deep learning model based on movement patterns can get over the limitations caused by the single authentication scenario, and achieve accuracy of 99.6%. Meanwhile, comparing with the CNN model alone, the accuracy of the proposed CNN-SACA model is improved by 1.5?percentage points, which effectively improves the authentication capability of mobile devices in multiple scenarios.

Key words: convolutional neural network, attention module, multi-scene, continuous authentication, mobile device

金瑜瑶, 张晓梅, 王亚杰. 基于注意力模块的移动设备多场景持续身份认证[J]. 计算机工程与应用, 2024, 60(3): 280-291.

JIN Yuyao, ZHANG Xiaomei, WANG Yajie. Multi-Scene Continuous Authentication Based on Attention Module for Mobile Devices[J]. Computer Engineering and Applications, 2024, 60(3): 280-291.

参考文献

[1] DATA.AI.2022年移动市场报告[EB/OL].[2022-06-11]. https://www.data.ai/cn/insights/market-data/state-of-mobile-2022/.
Data.ai.2022 Mobile market report[EB/OL].[2022-06-11]. https://www.data.ai/cn/insights/market-data/state-of-mobile-2022/.
[2] 厍向阳, 刘巧, 叶鸥.融合LeNet-5和Siamese神经网络模型的人脸认证算法研究[J]. 计算机工程与应用, 2020, 56(15): 215-220.
SHE X Y, LIU Q, YE O. Research on face verification algorithm based on LeNet-5 and Siamese neural network model[J]. Computer Engineering and Applications, 2020, 56(15): 215-220.
[3] 贺康, 李梦醒, 赵健, 等. 基于 Fingercode和同态加密的指纹认证方案[J].计算机工程与应用, 2013, 49(24): 78-82.
HE K, LI M X, ZHAO J, et al. Fingercode based remote fingerprint authentication scheme using homomorphic encryption[J]. Computer Engineering and Applications, 2013, 49(24): 78-82.
[4] ALROWAILY K, ALRUBAIAN M, MIRZA D A. Smart phones security-touch screen smudge attack[C]//Proceedings of the International Conference on Security and Management, 2012: 1-2.
[5] TSOKKIS P, STAVROU E. A password generator tool to increase users’ awareness on bad password construction strategies[C]//2018 International Symposium on Networks, Computers and Communications (ISNCC). Rome, Italy: IEEE, 2018: 1-5.
[6] LIU X, LI Y, DENG R H, et al. When human cognitive modeling meets PINs: user-independent inter-keystroke timing attacks[J]. Computers & Security, 2019, 80(1): 90-107.
[7] MARKERT P, BAILEY D V, GOLLA M, et al. This PIN can be easily guessed: analyzing the security of smartphone unlock PINs[C]//2020 IEEE Symposium on Security and Privacy (SP). San Francisco, CA, USA: IEEE, 2020: 286-303.
[8] SYED Z, HELMICK J, BANERJEE S, et al. Effect of user posture and device size on the performance of touch-based authentication systems[C]//IEEE International Symposium on High Assurance Systems Engineering. Daytona Beach Shores, FL, USA: IEEE, 2015: 10-17.
[9] WU J, ISHWAR P, KONRAD J. The value of posture, build and dynamics in gesture-based user authentication.[C]//IEEE International Joint Conference on Biometrics. Clearwater, FL, USA: IEEE, 2014: 1-8.
[10] TUTKUVIENE J, SCHIEFENHOEVEL W. Laterality of handgrip strength: age-and physical training-related changes in Lithuanian schoolchildren and conscripts[J]. Annals of the New York Academy of Sciences, 2013, 1288(1): 124-134.
[11] KUBOTA H, DEMURA S, KAWABATA H. Laterality and age-level differences between young women and elderly women in controlled force exertion (CFE)[J]. Archives of Gerontology and Geriatrics, 2012, 54(2): 68-72.
[12] TAHIR H, KHAN M S, TARIQ M O. Performance analysis and comparison of Faster R-CNN, Mask R-CNN and ResNet50 for the detection and counting of vehicles[C]//2021 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS).Greater Noida, India: IEEE, 2021: 587-594
[13] RAO G, ZHANG Y, ZHANG L, et al. MGL-CNN: a hierarchical posts representations model for identifying depressed individuals in online forums[J]. IEEE Access, 2020, 8(1): 32395-32403.
[14] LIN C Y, LI Y, ZHANG K, et al. CNN-based super resolution for video coding using decoded information[C]//International Conference on Visual Communications and Image Processing. Munich, Germany: IEEE, 2021: 1-5.
[15] MAYORGA P, VALDEZ J A, DRUZGALSKI C, et al. CNN networks to classify cardiopulmonary signals[C]//2022 Global Medical Engineering Physics Exchanges/Pan American Health Care Exchanges (GMEPE/PAHCE). Panama City, Panama: IEEE, 2022: 1-4.
[16] YANG Y, GUO B, WANG Z, et al. Behave sense: continuous authentication for security-sensitive mobile apps using behavioral biometrics[J]. Ad Hoc Networks, 2019, 84(3): 9-18.
[17] EL-SOUD M, GABER T, ALFAYEZ F, et al. Implicit authentication method for smartphone users based on rank aggregation and random forest[J]. Alexandria Engineering Journal, 2021, 60(1): 273-283.
[18] 芦效峰, 张胜飞, 伊胜伟. 基于CNN和RNN的自由文本击键模式持续身份认证[J]. 清华大学学报(自然科学版), 2018, 58(12): 1072-1078.
LU X F, ZHANG S F, YI S W. Free-text keystroke continuous authentication using CNN and RNN[J]. Journal of Tsinghua University (Science and Technology), 2018, 58(12): 1072-1078.
[19] SHIRAGA K, MAKIHARA Y, MURAMATSU D, et al. GEINet: view-invariant gait recognition using a convolutional neural network[C]//Proceedings of the 2016 International Conference on Biometrics (ICB). Halmstad, Sweden: IEEE, 2016: 1-8.
[20] MATTEO G, MICHELE R. IDNet: smartphone-based gait recognition with convolutional neural networks[J]. Pattern Recognition, 2018, 74(1): 25-37.
[21] ZHAO Y, ZHOU S. Wearable device-based gait recognition using angle embedded gait dynamic images and a convolutional neural network[J]. Sensors, 2017, 17(3): 1-20.
[22] TANG C, PHOHA V V. An empirical evaluation of activities and classifiers for user identification on smartphones[C]//2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS). Niagara Falls, NY, USA: IEEE, 2016: 1-8.
[23] 王欣, 王美丽, 边党伟. 融合MobileNetv2和注意力机制的轻量级人像分割算法[J].计算机工程与应用, 2022, 58(7): 220-228.
WANG X, WANG M L, BIAN D W. Algorithm for portrait segmentation combined with MobileNetv2 and attention mechanism[J]. Computer Engineering and Applications, 2022, 58(7): 220-228.
[24] 王玲敏, 段军, 辛立伟.引入注意力机制的YOLOv5安全帽佩戴检测方法[J].计算机工程与应用, 2022, 58(9): 303-312.
WANG L M, DUAN J, XIN L W. YOLOv5 helmet wear detection method with introduction of attention mechanism[J]. Computer Engineering and Applications, 2022, 58(9): 303-312.
[25] LI N, WANG Z. Spatial attention guided residual attention network for hyperspectral image classification[J]. IEEE Access, 2022, 10: 9830-9847.
[26] 杜先君, 巩彬, 余萍, 等. 基于CBAM-CNN的模拟电路故障诊断[J]. 控制与决策, 2022, 37(10): 2609-2618.
DU X J, GONG B, YU P, et al. CBAM-CNN based analog circuit fault diagnosis[J]. Control and Decision, 2022, 37(10): 2609-2618.
[27] 许文鑫, 张敏, 莫继良, 等. 基于CBAM-CNN的高速列车制动闸片摩擦块偏磨状态监控[J]. 摩擦学学报, 2022, 42(6): 1226-1236.
XU W X, ZHANG M, MO J L, et al. Monitoring of partial wear state of brake sluice friction block of high speed train based on CBAM-CNN[J]. Tribology, 2022, 42(6): 1226-1236.
[28] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision, Munich, Sep 8-14, 2018. Berlin, Heidelberg: Springer, 2018: 3-19.
[29] JIE H, LI S, GANG S, et al. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 7132-7141.
[30] YANG Q, PENG G, NGUYEN D T, et al. A multimodal data set for evaluating continuous authentication performance in smartphones[C]//Proceedings of the 12th ACM Conference on Embedded Network Sensor Systems. New York, NY, USA: Association for Computing Machinery, 2014: 358-359.
[31] BüCH H. Continuous authentication using inertial-sensors of smartphones and deep learning[D]. Stuttgart, Germany: University of Media Stuttgart, 2019.
[32] SITOVáZ, ?EDěNKA J, YANG Q, et al. HMOG: new behavioral biometric features for continuous authentication of smartphone users[J]. IEEE Transactions on Information Forensics and Security, 2016, 11(5): 877-892.
[33] CENTENO M P, GUAN Y, MOORSEL A V. Mobile based continuous authentication using deep features[C]//Proceedings of the 2nd International Workshop on Embedded and Mobile Deep Learning. New York, NY, USA: Association for Computing Machinery, 2018: 19-24.
[34] CHAO S. Performance analysis of multi-motion sensor behavior for active smartphone authentication[J]. IEEE Transactions on Information Forensics and Security, 2017, 13(1): 48-62.
[35] LI Y, HU H, GANG Z, et al. Sensor-based continuous authentication using cost-effective kernel ridge regression[J]. IEEE Access, 2018, 6: 32554-32565.
[36] LI Y, HU H, ZHOU G. Using data augmentation in continuous authentication on smartphones[J]. IEEE Internet of Things Journal, 2018, 6(1): 628-640.