Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Longitudinal symptom dynamics in postoperative patients with liver cancer: A cross-lagged panel network analysis.

Asia-Pacific journal of oncology nursing·2026
Same author

A rare case of massive upper gastrointestinal bleeding caused by Brunner's gland adenomas combined with a neuroendocrine tumor.

Endoscopy·2026
Same author

Barriers and facilitators of the implementation of a leadership training program based on the Developing and Sustaining Nursing Leadership Best Practice Guideline: A qualitative study.

International journal of nursing sciences·2026
Same author

A Prosthetic Hand System by Contralateral-Collaborative Control Based on Multi-task Learning.

IEEE journal of biomedical and health informatics·2026
Same author

Review on the Mechanism of and Therapies Targeting PANoptosis in Ulcerative Colitis.

Biomolecules·2026
Same author

Development of Multitaxon Indices of Biotic Integrity for Aquatic Ecosystem Health Assessment in Dongjiang Lake.

Biology·2026
Same journal

Turbulent flow in a vortex separator with a directed pipe inlet.

Scientific reports·2026
Same journal

Systematic characteristic evaluation of clay-based cementitious material derived from calcium carbide residue and waste tile powder.

Scientific reports·2026
Same journal

Retraction Note: Improvement of a rapid diagnostic application of monoclonal antibodies against avian influenza H7 subtype virus using Europium nanoparticles.

Scientific reports·2026
Same journal

Applying large language models to spam detection in the Kazakh low-resource language setting.

Scientific reports·2026
Same journal

An open-source 3D printing system enabling in-situ freeze-thaw processing of hydrogels.

Scientific reports·2026
Same journal

An enhanced EfficientNet framework for automated waste classification using cosine annealing and label smoothing.

Scientific reports·2026
See all related articles

Related Experiment Video

Updated: Jun 29, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

529

A study on expression recognition based on improved mobilenetV2 network.

Qiming Zhu1, Hongwei Zhuang2, Mi Zhao3

  • 1College of Equipment Support and Management, Engineering University of PAP, Xi'an, 710086, China.

Scientific Reports
|April 6, 2024
PubMed
Summary
This summary is machine-generated.

This study introduces an improved MobileNetV2 (I-MobileNetV2) for facial emotion recognition, significantly reducing parameters while enhancing accuracy. The enhanced model addresses feature loss and improves real-time performance for better emotion detection.

Keywords:
Attention mechanismExpression recognitionMobileNetV2Reverse fusionSELU

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.1K
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.1K

Related Experiment Videos

Last Updated: Jun 29, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

529
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.1K
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.1K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Deep convolutional neural networks often have large parameter quantities.
  • Lightweight neural networks like MobileNetV2 can suffer from feature information loss, poor real-time performance, and low accuracy in facial emotion recognition.

Purpose of the Study:

  • To propose an improved MobileNetV2 (I-MobileNetV2) strategy for facial emotion recognition.
  • To address the limitations of existing models, including large parameter counts and feature information loss.

Main Methods:

  • Inherited depthwise separable convolution from MobileNetV2 for reduced computational load.
  • Incorporated a reverse fusion mechanism to retain negative features and minimize information loss.
  • Replaced the RELU6 activation function with SELU to prevent gradient vanishing.
  • Integrated the Squeeze-and-Excitation Networks (SE-Net) channel attention mechanism to boost feature recognition.

Main Results:

  • The I-MobileNetV2 model achieved accuracies of 68.62% on FER2013 and 95.96% on CK+ datasets.
  • Accuracy improved by 0.72% (FER2013) and 6.14% (CK+) compared to the original MobileNetV2.
  • The parameter count was reduced by 83.8%, indicating a more lightweight model.

Conclusions:

  • The proposed I-MobileNetV2 model demonstrates significant improvements in facial emotion recognition accuracy and efficiency.
  • The integration of reverse fusion, SELU activation, and SE-Net effectively enhances feature recognition and reduces model complexity.