Towards Adversarially Robust Deep Image Denoising

by   Hanshu Yan, et al.
National University of Singapore

This work systematically investigates the adversarial robustness of deep image denoisers (DIDs), i.e, how well DIDs can recover the ground truth from noisy observations degraded by adversarial perturbations. Firstly, to evaluate DIDs' robustness, we propose a novel adversarial attack, namely Observation-based Zero-mean Attack (ObsAtk), to craft adversarial zero-mean perturbations on given noisy images. We find that existing DIDs are vulnerable to the adversarial noise generated by ObsAtk. Secondly, to robustify DIDs, we propose an adversarial training strategy, hybrid adversarial training (HAT), that jointly trains DIDs with adversarial and non-adversarial noisy data to ensure that the reconstruction quality is high and the denoisers around non-adversarial data are locally smooth. The resultant DIDs can effectively remove various types of synthetic and adversarial noise. We also uncover that the robustness of DIDs benefits their generalization capability on unseen real-world noise. Indeed, HAT-trained DIDs can recover high-quality clean images from real-world noise even without training on real noisy data. Extensive experiments on benchmark datasets, including Set68, PolyU, and SIDD, corroborate the effectiveness of ObsAtk and HAT.


page 4

page 5

page 12


Towards Adversarial Robustness of Deep Vision Algorithms

Deep learning methods have achieved great success in solving computer vi...

Improving Global Adversarial Robustness Generalization With Adversarially Trained GAN

Convolutional neural networks (CNNs) have achieved beyond human-level ac...

Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training

Existing deep learning real denoising methods require a large amount of ...

Hiding Images into Images with Real-world Robustness

The existing image embedding networks are basically vulnerable to malici...

Automatic Feature Highlighting in Noisy RES Data With CycleGAN

Radio echo sounding (RES) is a common technique used in subsurface glaci...

Analysis and Improvement of Adversarial Training in DQN Agents With Adversarially-Guided Exploration (AGE)

This paper investigates the effectiveness of adversarial training in enh...

NoiLIn: Do Noisy Labels Always Hurt Adversarial Training?

Adversarial training (AT) based on minimax optimization is a popular lea...

Please sign up or login with your details

Forgot password? Click here to reset