Abstract

Despite their outstanding performance, convolutional deep neural networks (DNNs) are vulnerable to small adversarial perturbations. In this Letter, we introduce a novel approach to thwart adversarial attacks. We propose to employ compressive sensing (CS) to defend DNNs from adversarial attacks, and at the same time to encode the image, thus preventing counterattacks. We present computer simulations and optical experimental results of object classification in adversarial images captured with a CS single pixel camera.

© 2021 Optical Society of America

Compressive imaging for thwarting adversarial attacks on 3D point cloud classifiers

Vladislav Kravets, Bahram Javidi, and Adrian Stern
Opt. Express 29(26) 42726-42737 (2021)

Snapshot temporal compressive microscopy using an iterative algorithm with untrained neural networks

Mu Qiao, Xuan Liu, and Xin Yuan
Opt. Lett. 46(8) 1888-1891 (2021)

Dynamic-range compression scheme for digital hologram using a deep neural network

Tomoyoshi Shimobaba, David Blinder, Michal Makowski, Peter Schelkens, Yota Yamamoto, Ikuo Hoshi, Takashi Nishitsuji, Yutaka Endo, Takashi Kakue, and Tomoyoshi Ito
Opt. Lett. 44(12) 3038-3041 (2019)

DeepCubeNet: reconstruction of spectrally compressive sensed hyperspectral images with deep neural networks

Daniel Gedalin, Yaniv Oiknine, and Adrian Stern
Opt. Express 27(24) 35811-35822 (2019)

Broadband achromatic metalens design based on deep neural networks

Xipeng An, Yue Cao, Yunxuan Wei, Zhihao Zhou, Tie Hu, Xing Feng, Guangqiang He, Ming Zhao, and Zhenyu Yang
Opt. Lett. 46(16) 3881-3884 (2021)