Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Feb 19, 2026

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Deep neural networks for texture classification-A theoretical analysis.

Saikat Basu¹, Supratik Mukhopadhyay¹, Manohar Karki¹

¹Louisiana State University, Baton Rouge, LA, USA.

Neural Networks : the Official Journal of the International Neural Network Society

|November 11, 2017

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Artificial capillaries-on-a-chip with modular control over lumen size, architecture, and co-culture conditions.

Biofabrication·2026

Same author

A multimodal human protein embeddings database: DeepDrug Protein Embeddings Bank (DPEB).

NAR genomics and bioinformatics·2026

Same author

Reduced-order modeling of solute transport within physiologically realistic solid tumor microenvironment.

Frontiers in pharmacology·2026

Same author

Artificial capillaries-on-a-chip with modular control over lumen size, architecture, in situ modifications and co-culture conditions.

bioRxiv : the preprint server for biology·2026

Same author

Domain growth and aging in a phase separating binary fluid confined inside a nanopore.

Soft matter·2026

Same author

Ai derivation and exploration of antibiotic class spaces.

Journal of cheminformatics·2026

Same journal

A boundary-regularization-enhanced video anomaly detection network based on context-adaptive spatio-temporal conditional diffusion.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

MT<sup>2</sup>-CSD and LLM-CRAN: A new dataset and an LLM-based multi-semantic knowledge fusion model for conversational stance detection.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

TriAlignNet: A triple-path cross-modality alignment framework for multimodal time series forecasting.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026

See all related articles

Deep Neural Networks (DNNs) show promise for texture-based image classification. Hand-crafted features reduce dimensionality and excess error, with new VC dimension bounds derived for DNNs and related networks.

Area of Science:

Computer Vision
Machine Learning
Pattern Recognition

Background:

Texture features are crucial for discriminative representations in image classification.
Deep Neural Networks (DNNs) are powerful tools for image analysis.
Understanding the impact of feature dimensionality on DNN performance is essential.

Purpose of the Study:

To investigate the efficacy of DNNs for image classification tasks emphasizing texture features.
To analyze the role of hand-crafted feature extraction in reducing dimensionality and error rates.
To derive novel theoretical bounds for the Vapnik-Chervonenkis (VC) dimension of various neural network architectures.

Main Methods:

Derivation of feature space size for standard textural features.

Keywords:

Deep neural network Texture classification vc dimension

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Feb 19, 2026

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Application of Vapnik-Chervonenkis (VC) dimension theory.

Analysis of intrinsic dimension for texture-based datasets.

Derivation of upper bounds on VC dimension for Convolutional Neural Networks (CNNs), Dropout, and Dropconnect networks.

Calculation of mean distance from centroid in n-dimensional manifolds.

Main Results:

Hand-crafted feature extraction leads to low-dimensional representations, reducing excess error rates.
Novel upper bounds on the VC dimension for CNNs, Dropout, and Dropconnect networks are established.
Texture-based datasets are shown to be inherently higher dimensional than digit or object recognition datasets.
Relative Contrast diminishes as the dimensionality of the vector space increases.

Conclusions:

DNNs, especially when combined with dimensionality reduction techniques, offer effective solutions for texture-based image classification.
The derived VC dimension bounds provide theoretical insights into the generalization capabilities of different neural network architectures.
The inherent dimensionality of texture data poses challenges for neural network learning, impacting their ability to shatter datasets.