Abstract

In this paper, a modified probabilistic deep learning method is proposed to attack the double random phase encryption by modeling the conditional distribution of plaintext. The well-trained probabilistic model gives both predictions of plaintext and uncertainty quantification, the latter of which is first introduced to optical cryptanalysis. Predictions of the model are close to real plaintexts, showing the success of the proposed model. Uncertainty quantification reveals the level of reliability of each pixel in the prediction of plaintext without ground truth. Subsequent simulation experiments demonstrate that uncertainty quantification can effectively identify poor-quality predictions to avoid the risk of unreliability from deep learning models.

© 2022 Optica Publishing Group

Untrained neural network for cryptanalysis of a phase-truncated-Fourier-transform-based optical cryptosystem

Shuixin Pan, Meihua Liao, Wenqi He, Yueqiang Zhang, and Xiang Peng
Opt. Express 29(26) 42642-42649 (2021)

Cryptanalysis of random-phase-encoding-based optical cryptosystem via deep learning

Han Hai, Shuixin Pan, Meihua Liao, Dajiang Lu, Wenqi He, and Xiang Peng
Opt. Express 27(15) 21204-21213 (2019)

Cryptanalysis of an “asymmetric optical cryptosystem based on coherent superposition and equal modulus decomposition”

Jingjing Wu, Wei Liu, Zhengjun Liu, and Shutian Liu
Appl. Opt. 54(30) 8921-8924 (2015)

Cryptographic analysis on an optical random-phase-encoding cryptosystem for complex targets based on physics-informed learning

Huazheng Wu, Qi Li, Xiangfeng Meng, Xiulun Yang, Shoupei Liu, and Yongkai Yin
Opt. Express 29(21) 33558-33571 (2021)

Known-plaintext cryptanalysis for a computational-ghost-imaging cryptosystem via the Pix2Pix generative adversarial network

Xiangru Liu, Xiangfeng Meng, Yurong Wang, Yongkai Yin, and Xiulun Yang
Opt. Express 29(26) 43860-43874 (2021)