Page 12 - IJEEE-2022-Vol18-ISSUE-1
P. 12
8 | Kadhim & Al-Darraji
VI. CONCLUSION [10] S. M. Moosavi-Dezfooli, A. Fawzi, and P. Frossard,
"Deepfool: a simple and accurate method to fool deep
To distinguish between original and adversarial neural networks," in Proceedings of the IEEE conference
images, a CNN algorithm was used. Therefore, ten classes on computer vision and pattern recognition, pp.
were used for ten people. That means each class for every 2574-2582, 2016.
person and everyone has many different captures of images
with different positions. Thus, classes were divided into [11] J. Xue, Y. Yang, and D. Jing, "Deceiving Face
groups of train and test datasets. So, every person has two Recognition Neural Network with Samples Generated by
folders for both train and test images. However, the number Deepfool," in Journal of Physics: Conference Series, vol.
of train images should be more than the images in the test to 1302, no. 2, 2019.
get clear results. Hence, the first method is by training
original images on the CNN algorithm. So, the rate of [12] Aleksander Madry, Aleksandar Makelov, Ludwig
recognition is "95%" between train images and test images. Schmidt, Dimitris Tsipras, Adrian Vladu, "Towards deep
learning models resistant to adversarial attacks,"
For future work, The number of attacks can be arXiv:1706.06083, 2017.
increased to distort images such as One Pixel attack, Carlini
& Wagner attacks (C&W), Visible Light-based attack [13] A. Goel, A. Agarwal, M. Vatsa, R. Singh, and N. K.
(VLA), AdvHat attack, Face Friend-safe attack, etc. The Ratha, "DNDNet: Reconfiguring CNN for adversarial
number of classes can also be increased with new images. robustness," in Proceedings of the IEEE/CVF Conference
on Computer Vision and Pattern Recognition Workshops,
CONFLICT OF INTEREST pp. 22-23, 2020.
The authors have no conflict of relevant interest to this [14] N. Carlini and D. Wagner, "Towards evaluating the
article. robustness of neural networks," in 2017 ieee symposium
on security and privacy (sp), pp. 39-57, 2017.
REFERENCES
[15] N. Papernot, P. McDaniel, X. Wu, S. Jha, and A.
[1] Mofeed T. Rashid, "Modeling Of Self-Organization Fish Swami, "Distillation as a defense to adversarial
School System By Neural Network System," Basrah perturbations against deep neural networks," in 2016 IEEE
Journal for Engineering Science, vol. 15, no. 1, pp. 14-19, symposium on security and privacy (SP), pp. 582-597,
2015. 2016.
[2] I. Goodfellow, Jonathon Shlens, Christian Szegedy, [16] D. Deb, J. Zhang, and A. K. Jain, "Advfaces:
"Explaining and harnessing adversarial examples," 2014. Adversarial face synthesis," in 2020 IEEE International
Joint Conference on Biometrics (IJCB), pp. 1-10, 2020.
[3] R. Huang, B. Xu, D. Schuurmans, and C. J. Szepesvári,
"Learning with a strong adversary," CoRR, 2015. [17] M. Cisse, P. Bojanowski, E. Grave, Y. Dauphin, and N.
Usunier, "Parseval networks: Improving robustness to
[4] Harini Kannan, Alexey Kurakin, Ian Goodfellow, adversarial examples," in International Conference on
"Adversarial logit pairing," CoRR, Vol. abs/1803.06373, Machine Learning, pp. 854-863, 2017.
2018.
[18] A. Dubey, L. Maaten, Z. Yalniz, Y. Li, and D. Mahajan,
[5] Alexey Kurakin, Ian Goodfellow, Samy Bengio, "Defense against adversarial images using web-scale
"Adversarial machine learning at scale," Computer Vision nearest-neighbor search," in Proceedings of the
and Pattern Recognition, 2016. IEEE/CVF Conference on Computer Vision and Pattern
Recognition, pp. 8767-8776, 2019.
[6] A. Bharati, R. Singh, M. Vatsa, K. W. Bowyer, and
Security, "Detecting facial retouching using supervised [19] G. B. Huang, M. Mattar, T. Berg, and E.
deep learning," IEEE Transactions on Information Learned-Miller, "Labeled faces in the wild: A database
Forensics and Security, vol. 11, no. 9, pp. 1903-1913, forstudying face recognition in unconstrained
2016. environments," in Workshop on faces in'Real-Life'Images:
detection, alignment, and recognition, 2008.
[7] Christian Szegedy, Wojciech Zaremba, Ilya Sutskever,
Joan Bruna, Dumitru Erhan, Ian Goodfellow, Rob Fergus,
"Intriguing properties of neural networks," Computer
Vision and Pattern Recognition, 2013.
[8] A. Agarwal, R. Singh, M. Vatsa, and N. Ratha, "Are
image-agnostic universal adversarial perturbations for
face recognition difficult to detect?," in 2018 IEEE 9th
International Conference on Biometrics Theory,
Applications and Systems (BTAS), pp. 1-7, 2018.
[9] Y. Liu, S. Mao, X. Mei, T. Yang, and X. Zhao,
"Sensitivity of Adversarial Perturbation in Fast Gradient
Sign Method," in 2019 IEEE Symposium Series on
Computational Intelligence (SSCI), pp. 433-436, 2019.