journal article·2022-02-15Open access

AT-BOD: An Adversarial Attack on Fool DNN-Based Blackbox Object Detection Models

Ilham A. Elaalami, Sunday O. Olatunji, Rachid M. Zagrouba

Applied Sciences· Volume 12· Issue 4

Abstract

Object recognition is a fundamental concept in computer vision. Object detection models have recently played a vital role in various applications, including real-time and safety-critical systems such as camera surveillance and self-driving cars. Scientific research has proven that object detection models are prone to adversarial attacks. Although several proposed methods exist throughout the literature, they either target white-box models or specific-task black-box models and do not generalize on other detectors. In this paper, we proposed a new adversarial attack against Blackbox-based object detectors called AT-BOD. The proposed AT-BOD model can fool the single-stage and multi-stage detectors, where we used an optimization algorithm to generate adversarial examples depending only on the detector predictions. AT-BOD model works in two diverse ways, reducing the confidence score and misleading the model to make the wrong decision or hide the object detection models. Our solution achieved a fooling rate of 97% and a false negative increase of 99% on the YOLOv3 detector, and a fooling rate of 61% false-negative increase of 57% on the Faster R-CNN detector. The detection accuracy of YOLOv3 and Faster R-CNN under AT-BOD was dramatically reduced and reached ≤1% and ≤3%, respectively.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Adversarial Robustness in Machine Learning · Advanced Neural Network Applications · Anomaly Detection Techniques and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

College of Computer Science and Information Technology, Imam Abdulrahman Bin Faisal University, P.O. Box 1982, Dammam 1441, Saudi Arabia

SAUDI ARAMCO Cybersecurity Chair, College of Computer Science and Information Technology, Imam Abdulrahman Bin Faisal University, P.O. Box 1982, Dammam 1441, Saudi Arabia

References (57)

Show source-provided references
1. Maqueda, A.I., Loquercio, A., Gallego, G., Garcia, N., and Scaramuzza, D. (2018, January 18–23). Event-Based Vision Meets Deep Learning on Steering Prediction for Self-Driving Cars. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00568 ↗
2. Chernikova, A., Oprea, A., Nita-Rotaru, C., and Kim, B. (2019, January 19–23). Are self-driving cars secure? Evasion attacks against deep neural networks for steering angle prediction. Proceedings of the IEEE Symposium on Security and Privacy Workshops, San Francisco, CA, USA.10.1109/SPW.2019.00033 ↗
3. Deep SCNN-Based real-time object detection for self-driving vehicles using LiDAR temporal data10.1109/ACCESS.2020.2990416 ↗
4. Sagduyu, Y.E., Shi, Y., and Erpek, T. (2019, January 1–9). IoT Network Security from the Perspective of Adversarial Deep Learning. Proceedings of the 2019 16th Annual IEEE International Conference on Sensing Communication, and Networking, Boston, MA, USA.10.1109/SAHCN.2019.8824956 ↗
5. Goodfellow, I.J., Shlens, J., and Szegedy, C. (2015, January 7–9). Explaining and harnessing adversarial examples. Proceedings of the 3rd International Conference on Learning Representations, ICLR 2015, Conference Track Proceedings, San Diego, CA, USA.
6. Cao, Y., Zhou, Y., Chen, Q.A., Xiao, C., Park, W., Fu, K., Cyr, B., Rampazzi, S., and Morley Mao, Z. (2019, January 9–12). Adversarial sensor attack on LiDAR-based perception in autonomous driving. Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security, Auckland, New Zealand.10.1145/3319535.3339815 ↗
7. Szegedy, C., Bruna, J., Erhan, D., and Goodfellow, I. (2014). Intriguing properties of neural networks. arXiv.
8. Cisse, M., Adi, Y., Neverova, N., and Keshet, J. (2017). Houdini: Fooling Deep Structured Prediction Models. arXiv.
9. Secure deep neural networks using adversarial image generation and training with Noise-GAN10.1016/j.cose.2019.06.012 ↗
10. Wang, B., Yao, Y., Shan, S., Li, H., Viswanath, B., Zheng, H., and Zhao, B.Y. (2019, January 19–23). Neural cleanse: Identifying and mitigating backdoor attacks in neural networks. Proceedings of the IEEE Symposium on Security and Privacy 2019, San Francisco, CA, USA.10.1109/SP.2019.00031 ↗
11. Security of emergent automotive systems: A tutorial introduction and perspectives on practice10.1109/MDAT.2019.2944086 ↗
12. Morgulis, N., Kreines, A., Mendelowitz, S., and Weisglass, Y. (2019). Fooling a Real Car with Adversarial Traffic Signs. arXiv.
13. Sitawarin, C., Bhagoji, A.N., Mosenia, A., Chiang, M., and Mittal, P. (2019). DARTS: Deceiving Autonomous Cars with Toxic Signs. arXiv.
14. Evasion attacks against machine learning at test time
15. Recent Advances in Deep Learning for Object Detection10.1016/j.neucom.2020.01.085 ↗
16. Melis, M., Demontis, A., Biggio, B., Brown, G., Fumera, G., and Roli, F. (2017, January 22–29). Is Deep Learning Safe for Robot Vision? Adversarial Examples against the iCub Humanoid. Proceedings of the IEEE International Conference on Computer Vision Workshops, Venice, Italy.10.1109/ICCVW.2017.94 ↗
17. Lu, J., Sibai, H., and Fabry, E. (2017). Adversarial Examples that Fool Detectors. arXiv.
18. Xie, C., Wang, J., Zhang, Z., Zhou, Y., Xie, L., and Yuille, A. (2017, January 22–29). Adversarial Examples for Semantic Segmentation and Object Detection. Proceedings of the IEEE International Conference on Computer Vision Workshops 2017, Venice, Italy.10.1109/ICCV.2017.153 ↗
19. Generative Adversarial Attacks Against Deep-Learning-Based Camera Model Identification
20. Adversarial attack-based security vulnerability verification using deep learning library for multimedia video surveillance10.1007/s11042-019-7262-8 ↗
21. Makelov, A., Schmidt, L., Tsipras, D., and Vladu, A. (2019). Towards Deep Learning Models Resistant to Adversarial Attacks. arXiv.
22. Fawzi, A., and Frossard, P. (2016, January 27–30). DeepFool: A simple and accurate method to fool deep neural networks. Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, NV, USA.
23. Carlini, N., and Wagner, D. (2017, January 22–26). Towards Evaluating the Robustness of Neural Networks. Proceedings of the 2017 IEEE Symposium on Security and Privacy, San Jose, CA, USA.10.1109/SP.2017.49 ↗
24. Chen, X., Liu, S., Xu, K., and Li, X. (2019). ZO-AdaMM: Zeroth-Order Adaptive Momentum Method for Black-Box Optimization. arXiv.
25. Thys, S., Ranst, W.V., and Goedeme, T. (2019, January 16–20). Fooling automated surveillance cameras: Adversarial patches to attack person detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA.10.1109/CVPRW.2019.00012 ↗
26. Rakin, A.S., and Fan, D. (2019, January 15–17). Defense-Net: Defend Against a Wide Range of Adversarial Attacks through Adversarial Detector. Proceedings of the 2019 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), Miami, FL, USA.10.1109/ISVLSI.2019.00067 ↗
27. Balakrishnan, A., Puranic, A.G., Qin, X., Dokhanchi, A., Deshmukh, J.V., Ben Amor, H., Fainekos, G., Yuan, X., He, P., and Zhu, Q. (2019, January 6–9). Adversarial Examples: Attacks and Defenses for Deep Learning. Proceedings of the 7th International Conference on Learning Representations. New Orleans, Louisiana, USA.
28. Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z.B., and Swami, A. (2017, January 2–6). Practical black-box attacks against machine learning. Proceedings of the ACM Asia Conference on Computer and Communications Security. Abu Dhabi, United Arab Emirates.10.1145/3052973.3053009 ↗
29. Liu, Y., Chen, X., Liu, C., and Song, D. (2017). Delving into Transferable Adversarial Examples and Black-Box Attacks. arXiv.
30. Narodytska, N. (2016). Simple Black-Box Adversarial Perturbations for Deep Networks. arXiv.10.1109/CVPRW.2017.172 ↗
31. Bhagoji, A.N., He, W., Li, B., and Song, D. (2018, January 8–14). Practical Black-box Attacks on Deep Neural Networks using Efficient Query Mechanisms. Proceedings of the 15th European Conference, Munich, Germany.10.1007/978-3-030-01258-8_10 ↗
32. Chen, P. (2017). ZOO: Zeroth Order Optimization Based Black-box Attacks to Deep Neural Networks without Training Substitute Models. arXiv.10.1145/3128572.3140448 ↗
33. Sharif, M., Bauer, L., and Reiter, M.K. (2016, January 24–28). Accessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face Recognition. Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, Vienna Austria.10.1145/2976749.2978392 ↗
34. Liu, Y., Lee, W., and Zhai, J. (2017). Trojaning Attack on Neural Networks, Purdue E-Pubs.10.14722/ndss.2018.23291 ↗
35. Brown, T.B., Mané, D., Roy, A., Abadi, M., and Gilmer, J. (2017, January 4–9). Adversarial Patch. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA.
36. Making the case for safety of machine learning in highly automated driving
37. Adversarial Attacks and Defenses in a Self-driving Car Environment
38. Geiger, A., Lenz, P., and Urtasun, R. (2012, January 16–21). Are we ready for autonomous driving? the KITTI vision benchmark suite. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Providence, RI, USA.10.1109/CVPR.2012.6248074 ↗
39. Eykholt, K., Evtimov, I., Fernandes, E., Li, B., Rahmati, A., Tram, F., Prakash, A., Kohno, T., and Song, D. (2018). Physical Adversarial Examples for Object Detectors. arXiv.
40. Machine learning, social learning and the governance of self-driving cars10.1177/0306312717741687 ↗
41. Lu, J., Sibai, H., Fabry, E., and Forsyth, D. (2017). NO Need to Worry about Adversarial Examples in Object Detection in Autonomous Vehicles. arXiv.
42. Mimouna, A., Alouani, I., Ben Khalifa, A., Hillali, Y.E., Taleb-Ahmed, A., Menhaj, A., Ouahabi, A., Essoukri, N., and Amara, B. (2020). electronics OLIMP: A Heterogeneous Multimodal Dataset for Advanced Environment Perception. Electronics, 9.10.3390/electronics9040560 ↗
43. Wang, X., Shrivastava, A., and Gupta, A. (2017, January 21–26). A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection. Proceedings of the Proceedings-30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA.10.1109/CVPR.2017.324 ↗
44. Lu, J., Sibai, H., Fabry, E., and Forsyth, D. (2017). Standard detectors aren’t (currently) fooled by physical adversarial stop signs. arXiv.
45. Models, P., and Li, Y. (2019). Robust Adversarial Perturbation on Deep Proposal-based Models. arXiv.
46. Chen, S.-T., Cornelius, C., Martin, J., and Horng Chau, D. (2018). ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector, Springer.10.1007/978-3-030-10925-7_4 ↗
47. Liu, X., Yang, H., Liu, Z., Song, L., Li, H., and Chen, Y. (2017). Dpatch: An adversarial patch attack on object detectors. arXiv.
48. Zeroth-Order Optimization and Its Application to Adversarial Machine Learning
49. Moosavi-Dezfooli, S.M., Fawzi, A., Fawzi, O., and Frossard, P. (2017, January 21–26). Universal adversarial perturbations. Proceedings of the 30th IEEE Computer Vision And Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.17 ↗
50. A Comprehensive Linear Speedup Analysis for Asynchronous Stochastic Parallel Optimization from Zeroth-Order to First-Order
51. Liu, S., Chen, P.Y., Chen, X., and Hong, M. (2019, January 6–9). SignsGD via zeroth-order oracle. Proceedings of the 7th International Conference on Learning Representations, New Orleans, LA, USA.
52. Kingma, D.P., and Lei Ba, J. (2014). Adam: A Method For Stochastic Optimization. arXiv.
53. Reddi, S.J., Kale, S., and Kumar, S. (2019). On the convergence of adam and beyond. arXiv.
54. The pascal visual object classes (VOC) challenge10.1007/s11263-009-0275-4 ↗
55. Dai, J., Li, Y., He, K., and Sun, J. (2016, January 5–10). R-FCN: Object detection via region-based fully convolutional networks. Proceedings of the Advances in Neural Information Processing Systems, Barcelona, Spain.
56. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks10.1109/TPAMI.2016.2577031 ↗
57. Deep Residual Learning for Image Recognition10.1016/0141-0229(95)00188-3 ↗