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

Coordination-Driven Metal-Phenolic Network Complexes for Programmable Colorimetric Signal Transduction and Defined Nano-Bio Interfaces.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Corrigendum to "Associations between multiple metals exposure and cognitive function in the middle-aged and older adults from China: A cross-sectional study" [Environ. Res., 263(Pt 1), 2024, 120038].

Environmental research·2026
Same author

SPP1 promotes lymph node metastasis and malignant progression of head and neck squamous cell carcinoma by activating FAK-AKT signaling pathway.

World journal of surgical oncology·2026
Same author

A retrospective cohort study comparing the effects on inflammatory response and clinical outcomes of vancomycin-PMMA versus vancomycin-sulfate calcium in chronic osteomyelitis.

Annals of medicine·2026
Same author

Author Correction: Parishin A-loaded mesoporous silica nanoparticles modulate macrophage polarization to attenuate tendinopathy.

NPJ Regenerative medicine·2026
Same author

Nanozymes-Armed Probiotic <i>Lactobacillus plantarum</i> Coordinates Ferroptosis-like Bacterial Death for Diabetic Wound Therapy.

Research (Washington, D.C.)·2026

Related Experiment Video

Updated: Jun 30, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.8K

Small object detection algorithm incorporating swin transformer for tea buds.

Meiling Shi1, Dongling Zheng1, Tianhao Wu2

  • 1College of Information Engineering, Sichuan Agricultural University, Ya'an, China.

Plos One
|March 21, 2024
PubMed
Summary

Accurate tea bud identification for robotic harvesting is improved by the STF-YOLO algorithm, which enhances feature extraction for better tea quality and yield. This novel approach significantly boosts small object detection performance in complex environments.

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

395
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.4K

Related Experiment Videos

Last Updated: Jun 30, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

395
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.4K

Area of Science:

  • Computer Vision
  • Robotics
  • Agricultural Technology

Background:

  • Accurate identification of small tea buds is crucial for automated tea harvesting robots, impacting tea quality and yield.
  • Existing image processing and machine learning methods struggle with the complexity and diversity of tea buds, leading to low accuracy and robustness.
  • Subtle features and morphology of small tea buds present significant challenges for current detection algorithms.

Purpose of the Study:

  • To propose a novel small object detection algorithm, STF-YOLO (Small Target Detection with Swin Transformer and Focused YOLO), for accurate identification of small tea buds.
  • To enhance the feature representation and detection capabilities for small objects in complex environments.
  • To improve the accuracy and robustness of tea bud identification for robotic harvesting applications.

Main Methods:

  • Integration of the Swin Transformer module for advanced visual feature extraction using self-attention mechanisms.
  • Leveraging the YOLOv8 network architecture, incorporating Focus and Depthwise Convolution modules for computational efficiency and improved feature handling.
  • Utilizing Wise Intersection over Union loss function for network optimization.
  • Development and experimentation on a self-created dataset of tea buds.

Main Results:

  • The STF-YOLO model achieved an accuracy of 91.5% and a mean Average Precision (mAP) of 89.4% on the tea bud dataset.
  • Demonstrated significant improvements over mainstream algorithms (YOLOv8, YOLOv7, YOLOv5, YOLOx), with accuracy gains of 5-20.22 percentage points and F1 score improvements of 0.03-0.13.
  • The proposed algorithm proved effective in enhancing small object detection performance in complex agricultural environments.

Conclusions:

  • STF-YOLO offers a robust and accurate solution for identifying small tea buds, addressing limitations of existing methods.
  • The research provides essential technical means for advancing automated tea harvesting and offers insights into small object detection challenges.
  • Future work can focus on further model optimization, data augmentation, and model fusion to enhance generalization and robustness for diverse scenarios.