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

Vision-Based Quality Grading of Beef Steaks Using Marbling Distribution Analysis and Lean Meat Color Classification.

Sensors (Basel, Switzerland)·2026
Same author

An Automated Vision-Based Inspection System for Metallic Lock Surface Defects Using a Transformer-Enhanced U-Net.

Sensors (Basel, Switzerland)·2026
Same author

Reprogramming Imine Reductases for Enantioselective Reduction of Electron-Rich Enamides.

JACS Au·2026
Same author

Smart Image-Based Deep Learning System for Automated Quality Grading of Phalaenopsis Seedlings in Outsourced Production.

Sensors (Basel, Switzerland)·2025
Same author

Ultrasensitive Direct Chemical Analysis of Human Hair Using Proton Transfer Reaction Time-of-Flight Mass Spectrometry (PTR-TOF-MS) for Nontargeted Exposure Profiling.

Chemical research in toxicology·2025
Same author

Smartphone-Based Sensing System for Identifying Artificially Marbled Beef Using Texture and Color Analysis to Enhance Food Safety.

Sensors (Basel, Switzerland)·2025

Related Experiment Video

Updated: Jan 13, 2026

Combination of Adhesive-tape-based Sampling and Fluorescence in situ Hybridization for Rapid Detection of Salmonella on Fresh Produce
09:10

Combination of Adhesive-tape-based Sampling and Fluorescence in situ Hybridization for Rapid Detection of Salmonella on Fresh Produce

Published on: October 18, 2010

13.4K

A Deep Learning-Based Sensing System for Identifying Salmon and Rainbow Trout Meat and Grading Freshness for Consumer

Hong-Dar Lin1, Jun-Liang Chen1, Chou-Hsien Lin2

  • 1Department of Industrial Engineering and Management, Chaoyang University of Technology, Taichung 413310, Taiwan.

Sensors (Basel, Switzerland)
|October 29, 2025
PubMed
Summary

This study introduces a smartphone-based system using deep learning to combat seafood fraud and assess salmon freshness. It empowers consumers with real-time fish identification and quality grading, enhancing food safety.

Keywords:
DenseNet121deep learningfish meat classificationfood fraudfreshness gradingtransfer learning

More Related Videos

DNA-based Fish Species Identification Protocol
09:15

DNA-based Fish Species Identification Protocol

Published on: April 28, 2010

30.1K
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

13.0K

Related Experiment Videos

Last Updated: Jan 13, 2026

Combination of Adhesive-tape-based Sampling and Fluorescence in situ Hybridization for Rapid Detection of Salmonella on Fresh Produce
09:10

Combination of Adhesive-tape-based Sampling and Fluorescence in situ Hybridization for Rapid Detection of Salmonella on Fresh Produce

Published on: October 18, 2010

13.4K
DNA-based Fish Species Identification Protocol
09:15

DNA-based Fish Species Identification Protocol

Published on: April 28, 2010

30.1K
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

13.0K

Area of Science:

  • Food Science
  • Computer Science
  • Artificial Intelligence

Background:

  • Seafood fraud, like mislabeling fish, presents significant food safety concerns and violates consumer rights.
  • Accurate identification and freshness assessment of seafood are crucial for consumer trust and safety.

Purpose of the Study:

  • To develop a deep learning-based, smartphone-compatible system for fish meat identification and salmon freshness grading.
  • To provide consumers with a real-time, image-based tool for verifying seafood authenticity and quality.

Main Methods:

  • A two-stage deep learning approach was employed, first classifying fish species and then grading salmon freshness.
  • An improved DenseNet121 architecture with global average pooling, dropout, and a custom output layer was utilized.
  • Transfer learning with partial layer freezing was applied to optimize training efficiency.

Main Results:

  • The two-stage method demonstrated superior performance compared to one-stage approaches and baseline models in classification and grading.
  • The system achieved robust accuracy in identifying fish types and assessing salmon freshness levels.
  • Sensitivity analysis indicated resilience to image distortions like blur and camera tilt.

Conclusions:

  • The developed system offers a practical, consumer-focused solution for seafood authentication and freshness evaluation.
  • This technology has the potential to significantly enhance food safety and protect consumer rights in the seafood market.
  • Further research is needed to address real-world challenges such as variable lighting and packaging conditions.