Using Deep learning for image watermarking attack

作者：

Highlights：

• Fully Convolutional Neural Network improves the denoising performance.

• Fully Convolutional Neural Network could be considered a deleterious attack.

• Fully Convolutional Neural Network preserves a good quality of the attacked images.

• Fully Convolutional Neural Network outperforms the other types of attacks.

摘要

•Fully Convolutional Neural Network improves the denoising performance.•Fully Convolutional Neural Network could be considered a deleterious attack.•Fully Convolutional Neural Network preserves a good quality of the attacked images.•Fully Convolutional Neural Network outperforms the other types of attacks.

论文关键词：Image denoising,Deep learning,Digital watermarking,Robustness,Fidelity

论文评审过程：Received 27 November 2019, Revised 29 June 2020, Accepted 6 October 2020, Available online 7 October 2020, Version of Record 8 October 2020.

论文官网地址：https://doi.org/10.1016/j.image.2020.116019