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 Experiment Videos

YOLOv12-VSD: A Transfer-Learning-Assisted Real-Time Detection Algorithm for Vehicle Surface Defects.

Haopu Liu1, Dequn Zhao2, Yu Li2

  • 1Beijing Dublin International College, Beijing University of Technology, Beijing 100124, China.

Sensors (Basel, Switzerland)
|May 13, 2026
PubMed
Summary

Related Concept Videos

Rolling Resistance: Problem Solving01:17

Rolling Resistance: Problem Solving

Rolling resistance, also known as rolling friction, is the force that resists the motion of a rolling object, such as a wheel, tire, or ball, when it moves over a surface. It is caused by the deformation of the object and the surface in contact with each other, as well as other factors like internal friction, hysteresis, and energy losses within the materials. Rolling resistance opposes the object's motion, requiring additional energy to overcome it and maintain movement. In practical...
Differential Leveling01:12

Differential Leveling

Differential leveling is a precise method in surveying used to determine the elevation difference between two points. Its primary goal is to establish accurate vertical measurements to create level surfaces or grade lines critical for designing and constructing infrastructures such as roads, bridges, and buildings.The procedure for differential leveling begins with setting up and leveling the instrument at a point where the benchmark can be seen. The level rod is held on the benchmark (BM), and...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Single-Metal-Anchored 1D Mesoporous Channels to Enable Accelerated Redox Kinetics for Lithium-Sulfur Batteries.

Nano-micro letters·2026
Same author

Halogen-Driven Ion Transport Homogenization in 3D Hierarchical MOF for Ultrastable Solid-State Lithium Metal Batteries.

Angewandte Chemie (International ed. in English)·2025
Same author

Design and Implementation of a Ku-Band High-Precision Blackbody Calibration Target.

Micromachines·2023
Same author

Diagnosing Diabetic Retinopathy in OCTA Images Based on Multilevel Information Fusion Using a Deep Learning Framework.

Computational and mathematical methods in medicine·2022
Same author

A Machine Vision Approach for Classification of Skin Cancer Using Hybrid Texture Features.

Computational intelligence and neuroscience·2022

This study introduces YOLOv12-VSD, an advanced algorithm for vehicle surface defect detection. It significantly improves accuracy and efficiency by addressing classification-localization inconsistency and multi-scale feature challenges.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Industrial Automation

Background:

  • Vehicle surface defect detection is crucial for quality control.
  • Existing methods struggle with classification-localization inconsistency, multi-scale defects, and domain shift.

Purpose of the Study:

  • To propose YOLOv12-VSD, an improved detection algorithm for vehicle surface defects.
  • To enhance accuracy, efficiency, and generalization capabilities in defect detection systems.

Main Methods:

  • Introduced an IoU-aware classification loss for better localization.
  • Implemented P4-RepC3 for enriched intermediate features and P5-SPPF for expanded receptive fields.
  • Developed a three-stage transfer learning framework to mitigate domain shift.
Keywords:
IoU-aware classification lossYOLOv12deep learningobject detectionstructural reparameterizationtransfer learningvehicle surface defect

Related Experiment Videos

Main Results:

  • YOLOv12-VSD achieved the highest mAP@50 (0.715) among six models with 6.1M parameters and 17.1 GFLOPs.
  • Three-stage transfer learning boosted mAP@50 from 0.531 to 0.652, reducing training time by 64%.

Conclusions:

  • YOLOv12-VSD effectively addresses key challenges in vehicle surface defect detection.
  • The proposed methods offer a robust and efficient solution for industrial applications with limited annotations.