• Home
  • Search
  • A Structurally Regularized CNN Architecture via Adaptive Subband Decomposition.
  • Open Access IconOpen Access
  • Cite Icon3
  • https://doi.org/10.1109/tnnls.2024.3486181Copy DOI Icon

A Structurally Regularized CNN Architecture via Adaptive Subband Decomposition.

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

We propose a generalized convolutional neural network (CNN) architecture that first decomposes the input signal into subbands by an adaptive filter bank structure, and then uses convolutional layers to extract features from each subband independently. Fully connected (FC) layers finally combine the extracted features to perform classification. The proposed architecture restrains each of the subband CNNs from learning using the entire input signal spectrum, resulting in structural regularization. Our proposed CNN architecture is fully compatible with the end-to-end learning mechanism of typical CNN architectures and learns the subband decomposition from the input dataset. We show that the proposed CNN architecture has attractive properties, such as robustness to input and weight-and-bias quantization noise, compared to regular full-band CNN architectures. Importantly, the proposed architecture significantly reduces computational costs, while maintaining state-of-the-art classification accuracy. Experiments on image classification tasks using the MNIST, CIFAR-10/100, Caltech-101, and ImageNet-2012 datasets show that the proposed architecture allows accuracy surpassing state-of-the-art results. On the ImageNet-2012 dataset, we achieved top-5 and top-1 validation set accuracy of 86.91% and 69.73%, respectively. Notably, the proposed architecture offers over 90% reduction in computation cost in the inference path and approximately 75% reduction in back-propagation (per iteration) with just a single-layer subband decomposition. With a two-layer subband decomposition, the computational gains are even more significant with comparable accuracy results to the single-layer decomposition.

Similar Papers
  • Dissertation

Evolving Convolutional Neural Networks for Image and Text Classification Tasks

  • Nov 24, 2024
  • Trevor Londt
  • Supplementary Content

Texture Patterns for Object Recognition and Content-Based Color Image Retrieval

  • Dec 21, 2020
  • Griffith Research Online (Griffith University, Queensland, Australia)
  • Muhammad Asif
  • Research Article
  • Citations7

A novel study for automatic two-class COVID-19 diagnosis (between COVID-19 and Healthy, Pneumonia) on X-ray images using texture analysis and 2-D/3-D convolutional neural networks

  • Jan 29, 2022
  • Multimedia Systems
  • Huseyin Ya\U015Far +1
  • Research Article
  • Citations4

Human Activity Recognition in a Realistic and Multiview Environment Based on Two-Dimensional Convolutional Neural Network

  • May 09, 2023
  • Journal of Artificial Intelligence and Technology
  • Ashish Khare +2
  • Conference Article
  • Citations3

Face Recognition with Multi-channel Local Mesh High-order Pattern Descriptor and Convolutional Neural Network

  • Dec 01, 2018
  • M Daud Abdullah Asif +2
  • Research Article
  • Citations885

Automatically Designing CNN Architectures Using the Genetic Algorithm for Image Classification.

  • Apr 21, 2020
  • IEEE Transactions on Cybernetics
  • Yanan Sun +4
  • Conference Article
  • Citations3

A Structurally Regularized Convolutional Neural Network for Image Classification Using Wavelet-Based Subband Decomposition

  • Sep 01, 2019
  • Pavel Sinha +2
  • Research Article
  • Citations72

Hardware implementation of 1D-CNN architecture for ECG arrhythmia classification

  • Mar 20, 2023
  • Biomedical Signal Processing and Control
  • Viraj Rawal +2
  • Conference Article
  • Citations66

Particle Swarm optimisation for Evolving Deep Neural Networks for Image Classification by Evolving and Stacking Transferable Blocks

  • Jul 01, 2020
  • Bin Wang +2
  • Conference Article
  • Citations9

Development of CNN architecture for Honey Bees Disease Condition

  • Sep 01, 2019
  • 2019 International Artificial Intelligence and Data Processing Symposium (IDAP)
  • Hüseyin Üzen +2
  • Research Article
  • Citations6

Analysis of CNN Architectures for Human Action Recognition in Video

  • Jun 30, 2022
  • Computación y Sistemas
  • David Silva +4
  • Research Article
  • Citations50

Receptive Field Regularization Techniques for Audio Classification and Tagging With Deep Convolutional Neural Networks

  • Jan 01, 2021
  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • Khaled Koutini +2
  • Conference Article
  • Citations2

Effective Deep Learning-based Side-Channel Analyses Against ASCAD

  • Oct 01, 2021
  • Junkai Liu +2
  • PDF
  • Research Article
  • Citations25

Is it enough to optimize CNN architectures on ImageNet?

  • Nov 15, 2022
  • Frontiers in Computer Science
  • Lukas Tuggener +2
  • Conference Article
  • Citations17

Classification of Hand-Drawn Basic Circuit Components Using Convolutional Neural Networks

  • Jun 01, 2020
  • 2020 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)
  • Mihriban Gunay +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.