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

Related Concept Videos

Force Classification

Force Classification

Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...

You might also read

Related Articles

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

Sort by

Same author

Development of ultra-high efficiency soft x-ray angle-resolved photoemission spectroscopy equipped with deep prior-based denoising method.

The Review of scientific instruments·2026

Same author

LLaVA-Pose: Keypoint-Integrated Instruction Tuning for Human Pose and Action Understanding.

Sensors (Basel, Switzerland)·2025

Same author

Exploring the role of texture features in deep convolutional neural networks: Insights from Portilla-Simoncelli statistics.

Neural networks : the official journal of the International Neural Network Society·2023

Same author

Editorial for "A Lightweight Convolutional Neural Network Based on Dynamic Level-Set Loss Function for Spine MR Image Segmentation".

Journal of magnetic resonance imaging : JMRI·2023

Same author

Development of spectral decomposition based on Bayesian information criterion with estimation of confidence interval.

Science and technology of advanced materials·2020

Same author

Anatomical and embryological perspectives in laparoscopic complete mesocoloic excision of splenic flexure cancers.

Surgical endoscopy·2017

Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

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

Same journal

New results on prescribed-time synchronization of complex networks via intermittent control.

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

Same journal

Variance-constrained multi-view ensemble broad network for imbalanced data.

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

Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

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

Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

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

Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

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

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 18, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Distorted image classification using neural activation pattern matching loss.

Satoshi Suzuki¹, Shoichiro Takeda², Ryuichi Tanida²

¹NTT Computer and Data Science Laboratories, NTT Corporation, 1-1 Hikarinooka, Yokosuka, 2390847, Kanagawa, Japan.

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

|August 25, 2023

Summary

This summary is machine-generated.

This study introduces a novel neural activation pattern matching (NAPM) loss to improve deep neural network (DNN) accuracy for distorted image classification. The NAPM loss simplifies the decision boundary, enabling more efficient optimization and higher accuracy on noisy or blurry images.

Keywords:

Deep neural network Distorted image classification Neural activation pattern matching loss

More Related Videos

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Published on: April 12, 2024

Cross-Modal Multivariate Pattern Analysis

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

Related Experiment Videos

Last Updated: Jul 18, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Published on: April 12, 2024

Cross-Modal Multivariate Pattern Analysis

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Deep neural networks (DNNs) trained on clean images struggle with distorted inputs like noise or blur, leading to reduced accuracy.
Existing methods retrain DNNs on both clean and distorted images, but often result in overly complex decision boundaries that hinder optimization.
A complex decision boundary limits the efficiency of DNNs in classifying images with varying quality.

Purpose of the Study:

To develop a novel loss function for improved distorted image classification.
To simplify the decision boundary in DNNs for more efficient optimization.
To enhance the accuracy of DNNs when classifying both distorted and undistorted images.

Main Methods:

Introduced a "neural activation pattern matching (NAPM) loss" function.
Leveraged the piecewise linear nature of DNN decision boundaries.
Matched neural activation patterns between distorted and undistorted images using sigmoid cross-entropy to constrain classification.

Main Results:

The NAPM loss constrains DNNs to classify distorted and undistorted images using the same decision boundary segment.
This approach accelerates optimization by preventing decision boundary over-complication.
Experimental results show increased accuracy compared to previous methods across all tested conditions.

Conclusions:

The NAPM loss offers a simple yet effective method for classifying distorted images.
It enhances DNN performance by ensuring a more regularized and efficient decision boundary.
This technique improves the robustness and accuracy of image classification systems facing image degradation.