mlsanityData to 14 Oct 2026

Frequency Centric Defense Mechanisms against Adversarial Examples

arXiv 2110.13935 ↗First posted 26 Oct 2021cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Adversarial example (AE) aims at fooling a Convolution Neural Network by introducing small perturbations in the input image.The proposed work uses the magnitude and phase of the Fourier Spectrum and the entropy of the image to defend against AE. We demonstrate the defense in two ways: by training an adversarial detector and denoising the adversarial effect. Experiments were conducted on the low-resolution CIFAR-10 and high-resolution ImageNet datasets. The adversarial detector has 99% accuracy for FGSM and PGD attacks on the CIFAR-10 dataset. However, the detection accuracy falls to 50% for sophisticated DeepFool and Carlini & Wagner attacks on ImageNet. We overcome the limitation by using autoencoder and show that 70% of AEs are correctly classified after denoising.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 17 of its 23 references that are in the map. First posted on arXiv 26 Oct 2021.

Builds on

17 of 23 references.

References · 17

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  6. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  7. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,318
  8. 2020On Adaptive Attacks to Adversarial Example Defensescited by 555
  9. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  10. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 282
  11. 2016Adversarial Diversity and Hard Positive Generationcited by 35
  12. 2017Adversarial Example Defenses: Ensembles of Weak Defenses are not Strongcited by 32
5 more
  1. 2018On the Structural Sensitivity of Deep Convolutional Networks to the Directions of Fourier Basis Functionscited by 32
  2. 2021SpectralDefense: Detecting Adversarial Attacks on CNNs in the Fourier Domaincited by 30
  3. 2019PuVAE: A Variational Autoencoder to Purify Adversarial Examplescited by 28
  4. 2017Denoising Adversarial Autoencoderscited by 20
  5. 2018Combatting Adversarial Attacks through Denoising and Dimensionality Reduction: A Cascaded Autoencoder Approachcited by 10

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 1.

3 papers.

  1. 2023Boosting Black-box Attack to Deep Neural Networks with Conditional Diffusion Modelscited by 12
  2. 2022Metricizing the Euclidean Space towards Desired Distance Relations in Point Cloudscited by 3
  3. 2024DFT-Based Adversarial Attack Detection in MRI Brain Imaging: Enhancing Diagnostic Accuracy in Alzheimer's Case Studiescited by 1