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

Artificial intelligence for detecting fetal orofacial clefts and advancing medical education.

Nature communications·2026
Same author

Analytical Evaluation of Stress-Strain Behavior and Reaction Mechanism of Lunar Regolith Simulant (CQU-1) Geopolymer.

Polymers·2026
Same author

Systematic Investigation of Microstructural and Spectral Characteristics in Citrus Midrib for Huanglongbing Detection.

Phytopathology·2026
Same author

Saliency-Aware Mutual Learning for enhanced retinal Image Quality Assessment in diabetic retinopathy diagnostics.

Experimental eye research·2026
Same author

Cysteine-Mediated Root Sequestration and Metabolic Reprogramming Alleviate Cadmium Toxicity in Chinese Cabbage.

Plant, cell & environment·2026
Same author

Knowledge, Attitudes, Practices, and Vaccination Willingness Toward Mpox (Monkeypox) Among Chinese Medical Students: Cross-Sectional Study.

JMIR public health and surveillance·2026

Related Experiment Video

Updated: Jun 4, 2025

Analysis of Raw and Processed Cyperi Rhizoma Samples Using Liquid Chromatography-Tandem Mass Spectrometry in Rats with Primary Dysmenorrhea
07:36

Analysis of Raw and Processed Cyperi Rhizoma Samples Using Liquid Chromatography-Tandem Mass Spectrometry in Rats with Primary Dysmenorrhea

Published on: December 23, 2022

1.5K

Improved CSW-YOLO Model for Bitter Melon Phenotype Detection.

Haobin Xu1, Xianhua Zhang1, Weilin Shen1

  • 1College of Horticulture, Fujian Agriculture and Forestry University, Fuzhou 350002, China.

Plants (Basel, Switzerland)
|December 17, 2024
PubMed
Summary

A new CSW-YOLO model improves bitter melon phenotype detection using advanced deep learning. This automated approach enhances accuracy and efficiency for crop breeding and agricultural technology.

Keywords:
CSW-YOLObitter melondeep learningphenotypic detection

More Related Videos

HPLC Coupled with Chemical Fingerprinting for Multi-Pattern Recognition for Identifying the Authenticity of Clematidis Armandii Caulis
07:29

HPLC Coupled with Chemical Fingerprinting for Multi-Pattern Recognition for Identifying the Authenticity of Clematidis Armandii Caulis

Published on: November 11, 2022

1.9K
Screening of Tobacco Genotypes for Phytophthora nicotianae Resistance
10:10

Screening of Tobacco Genotypes for Phytophthora nicotianae Resistance

Published on: April 15, 2022

3.3K

Related Experiment Videos

Last Updated: Jun 4, 2025

Analysis of Raw and Processed Cyperi Rhizoma Samples Using Liquid Chromatography-Tandem Mass Spectrometry in Rats with Primary Dysmenorrhea
07:36

Analysis of Raw and Processed Cyperi Rhizoma Samples Using Liquid Chromatography-Tandem Mass Spectrometry in Rats with Primary Dysmenorrhea

Published on: December 23, 2022

1.5K
HPLC Coupled with Chemical Fingerprinting for Multi-Pattern Recognition for Identifying the Authenticity of Clematidis Armandii Caulis
07:29

HPLC Coupled with Chemical Fingerprinting for Multi-Pattern Recognition for Identifying the Authenticity of Clematidis Armandii Caulis

Published on: November 11, 2022

1.9K
Screening of Tobacco Genotypes for Phytophthora nicotianae Resistance
10:10

Screening of Tobacco Genotypes for Phytophthora nicotianae Resistance

Published on: April 15, 2022

3.3K

Area of Science:

  • Agricultural Science
  • Computer Vision
  • Biotechnology

Background:

  • Bitter melon (Momordica charantia) is a valuable crop with growing market demand due to its medicinal and nutritional properties.
  • Accurate identification of bitter melon germplasm is vital for breeding programs, but traditional methods are slow and imprecise.
  • There is a need for automated and intelligent solutions for bitter melon phenotype detection to improve efficiency and accuracy.

Purpose of the Study:

  • To develop an automated and intelligent bitter melon phenotype detection model.
  • To enhance the accuracy and efficiency of bitter melon germplasm identification.
  • To provide technical support for visual detection technologies in agriculture.

Main Methods:

  • Developed a novel bitter melon phenotype detection model named CSW-YOLO.
  • Integrated the ConvNeXt V2 module into the YOLOv8 backbone for improved feature focus.
  • Incorporated the SimAM attention mechanism to enhance recognition accuracy without increasing parameters.
  • Utilized WIoUv3 as the bounding box loss function for faster convergence and better positioning.
  • Trained and tested the model on a comprehensive bitter melon image dataset.

Main Results:

  • The CSW-YOLO model achieved high performance metrics: 94.6% precision, 80.6% recall, 96.7% mAP50, and 87.04% F1 score.
  • Demonstrated significant improvements over the original YOLOv8 model in precision (8.5%), mAP50 (11.1%), and F1 score (4%).
  • Heatmap analysis and ablation studies confirmed enhanced target feature focus, reduced false detections, and improved generalization.
  • Comparative tests showed CSW-YOLO outperformed other mainstream deep learning models in bitter melon phenotype detection.

Conclusions:

  • The CSW-YOLO model offers an accurate and reliable method for bitter melon phenotype identification.
  • The developed model enhances automation and intelligence in agricultural phenotype detection.
  • This research provides valuable technical support for visual detection technologies in agriculture, applicable to other crops as well.