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

Machine learning-driven discovery of antimicrobial peptides against <i>Pseudomonas aeruginosa</i>.

Frontiers in pharmacology·2026
Same author

Machine Learning Algorithms Enabled Visual On-Site Intelligent Sensing of Bioactive Components by Histidine-Functionalized Nanozyme Sensor Array.

ACS sensors·2026
Same author

Spectroscopic Methods Applied in Food Quality Determination.

Foods (Basel, Switzerland)·2026
Same author

Outcomes of endolymphatic sac surgery for lermoyez syndrome: a 2-year and 10-year follow-up.

Acta oto-laryngologica·2026
Same author

MFAP4 activates PI3K/AKT-NF-κB signaling to drive tumor progression and immune evasion in bladder cancer.

International immunopharmacology·2026
Same author

Immune-mediated renal injury and cardiometabolic risk in IgA nephropathy: clinical evidence on telitacicept from a scoping review.

Frontiers in nutrition·2026

Related Experiment Video

Updated: Jun 13, 2026

Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements
10:25

Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements

Published on: June 28, 2016

High-Accuracy Prediction of Chunmee Tea Grade via DeepSpectra Model and Near-Infrared Spectroscopy.

Yatong Zhang1, Mobing Ren2, Xiaohong Wu3,4

  • 1Mengxi Honors College, Jiangsu University, Zhenjiang 212013, China.

Foods (Basel, Switzerland)
|June 12, 2026
PubMed
Summary

A new DeepSpectra model accurately grades Chunmee tea using near-infrared spectroscopy. This advanced method improves quality control and market valuation for green tea.

Keywords:
Chunmee teaconvolutional neural networkend-to-end learninginception modulelight gradient boosting machinemulti-scale feature extractionnear-infrared spectroscopy

More Related Videos

O-cresol Concentration Online Measurement Based On Near Infrared Spectroscopy Via Partial Least Square Regression
06:50

O-cresol Concentration Online Measurement Based On Near Infrared Spectroscopy Via Partial Least Square Regression

Published on: November 8, 2019

Tea Aroma Analysis Based on Solvent-Assisted Flavor Evaporation Enrichment
04:36

Tea Aroma Analysis Based on Solvent-Assisted Flavor Evaporation Enrichment

Published on: May 26, 2023

Related Experiment Videos

Last Updated: Jun 13, 2026

Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements
10:25

Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements

Published on: June 28, 2016

O-cresol Concentration Online Measurement Based On Near Infrared Spectroscopy Via Partial Least Square Regression
06:50

O-cresol Concentration Online Measurement Based On Near Infrared Spectroscopy Via Partial Least Square Regression

Published on: November 8, 2019

Tea Aroma Analysis Based on Solvent-Assisted Flavor Evaporation Enrichment
04:36

Tea Aroma Analysis Based on Solvent-Assisted Flavor Evaporation Enrichment

Published on: May 26, 2023

Area of Science:

  • Agricultural Science
  • Spectroscopy
  • Machine Learning

Background:

  • Chunmee tea quality grading is vital for market valuation.
  • Traditional methods and standard 1D-CNNs struggle with spectral information loss and multi-scale feature capture.

Purpose of the Study:

  • To develop an end-to-end DeepSpectra model for automatic Chunmee tea grading.
  • To overcome limitations of manual feature extraction and fixed kernels in traditional models.

Main Methods:

  • Collected 360 Chunmee tea samples across six grades using near-infrared spectroscopy.
  • Applied Multiplicative Scatter Correction (MSC) preprocessing.
  • Developed an improved DeepSpectra model with Inception module and residual connections.

Main Results:

  • Achieved an average test accuracy of 96.39 ± 1.63% across ten random divisions.
  • Significantly outperformed other models (p < 0.05).
  • Demonstrated excellent stability, generalization in small-sample scenarios, and low inference latency (2.2 ms).

Conclusions:

  • The proposed DeepSpectra model offers a reliable and efficient method for Chunmee tea grading.
  • Provides a promising strategy for intelligent and rapid quality control in the green tea industry.