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

Influence of food groups on plasma total homocysteine for specific MTHFR C677T genotypes in Chinese population.

Molecular nutrition & food research·2016
Same author

NIR Light Propulsive Janus-like Nanohybrids for Enhanced Photothermal Tumor Therapy.

Small (Weinheim an der Bergstrasse, Germany)·2016
Same author

Smart Hydrogels with Inhomogeneous Structures Assembled Using Nanoclay-Cross-Linked Hydrogel Subunits as Building Blocks.

ACS applied materials & interfaces·2016
Same author

Synergy between von Hippel-Lindau and P53 contributes to chemosensitivity of clear cell renal cell carcinoma.

Molecular medicine reports·2016
Same author

Aerobic Degradation of Sulfadiazine by Arthrobacter spp.: Kinetics, Pathways, and Genomic Characterization.

Environmental science & technology·2016
Same author

Downregulation of ClC-3 in dorsal root ganglia neurons contributes to mechanical hypersensitivity following peripheral nerve injury.

Neuropharmacology·2016

Related Experiment Video

Updated: Jul 24, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K

Surface-Framework structure: A neural network structure for weakening gridding effect in PCB mark-point semantic

Yeshuai Wang1, Jianhua Song2,3,4, Shihui Wang1

  • 1School of Computer Science and Information Engineering, Hubei University, Wuhan, Hubei, China.

Plos One
|July 10, 2023
PubMed
Summary

A novel surface-framework structure enhances image segmentation for PCB manufacturing. This pure efficient Unet (PE Unet) model improves accuracy and maintains speed, balancing performance and computational demands.

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.7K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

442

Related Experiment Videos

Last Updated: Jul 24, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.7K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

442

Area of Science:

  • Computer Vision
  • Machine Learning
  • Image Processing

Background:

  • Image transfer is critical for PCB manufacturing, impacting production speed and quality.
  • Existing semantic segmentation methods may face limitations in detail extraction and computational efficiency.

Purpose of the Study:

  • To propose a novel surface-framework structure for enhanced semantic segmentation.
  • To introduce the pure efficient Unet (PE Unet) model integrating this structure.
  • To evaluate the model's performance in PCB manufacturing and other datasets.

Main Methods:

  • A surface-framework network architecture is designed, separating surface and framework components.
  • The surface component avoids subsampling to preserve detailed image features.
  • The PE Unet model is developed based on Unet and the proposed structure, tested on MPRS, CHASE_DB1, and TCGA-LGG datasets.

Main Results:

  • PE Unet achieved an Intersection over Union (IoU) of 84.74% on the MPRS dataset, exceeding Unet by 3.15%.
  • The model demonstrates a balance between performance and speed with 34.0 GFLOPs.
  • Consistent IoU improvements were observed across MPRS (2.38%), CHASE_DB1 (4.35%), and TCGA-LGG (0.78%) datasets.

Conclusions:

  • The surface-framework structure effectively weakens the gridding effect in semantic segmentation.
  • PE Unet offers improved performance and efficiency for image segmentation tasks, particularly in PCB manufacturing.
  • The proposed method provides a viable solution for high-quality, high-speed image analysis in industrial applications.