Robust Image Classification Using a Low-Pass Activation Function and DCT Augmentation

Convolutional Neural Network's (CNN's) performance disparity on clean and corrupted datasets has recently come under scrutiny. In this work, we analyse common corruptions in the frequency domain, i.e., High Frequency corruptions (HFc, e.g., noise) and Low Frequency corruptions (LFc, e.g.,...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2021, Vol.9, p.86460-86474
Main Authors: Hossain, Md Tahmid, Teng, Shyh Wei, Sohel, Ferdous, Lu, Guojun
Format: Article
Language:English
Subjects:
activation function
Artificial neural networks
Augmentation
Benchmark testing
data augmentation
Discrete cosine transform
Discrete cosine transforms
Frequency-domain analysis
Hybrid fiber coaxial cables
Hydrofluorocarbons
Image classification
image corruption
input corruption
Low pass filters
low-pass filtering
Perturbation
Robust image classification
Robustness
Stability analysis
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Convolutional Neural Network's (CNN's) performance disparity on clean and corrupted datasets has recently come under scrutiny. In this work, we analyse common corruptions in the frequency domain, i.e., High Frequency corruptions (HFc, e.g., noise) and Low Frequency corruptions (LFc, e.g., blur). Although a simple solution to HFc is low-pass filtering, ReLU - a widely used Activation Function (AF), does not have any filtering mechanism. In this work, we instill low-pass filtering into the AF (LP-ReLU) to improve robustness against HFc. To deal with LFc, we complement LP-ReLU with Discrete Cosine Transform based augmentation. LP-ReLU, coupled with DCT augmentation, enables a deep network to tackle the entire spectrum of corruption. We use CIFAR-10-C and Tiny ImageNet-C for evaluation and demonstrate improvements of 5% and 7.3% in accuracy respectively, compared to the State-Of-The-Art (SOTA). We further evaluate our method's stability on a variety of perturbations in CIFAR-10-P and Tiny ImageNet-P, achieving new SOTA in these experiments as well. To further strengthen our understanding regarding CNN's lack of robustness, a decision space visualisation process is proposed and presented in this work.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2021.3089598