Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Long-Term Cognitive Impairment After CAR-T Therapy Versus Autologous Stem Cell Transplantation: A Propensity Score-Matched Cohort Study.

Diagnostics (Basel, Switzerland)·2026

Same author

Tiny Yeast Comet 1-Dependent Polymers Suppress Taxol-Stabilized Microtubule Depolymerization Induced by Ice-Cold CaCl<sub>2</sub>.

International journal of molecular sciences·2026

Same author

Sleep Stage Classification During CPAP Therapy from CPAP-Airflow and Wearable Fingertip Signals.

Sensors (Basel, Switzerland)·2026

Same author

COVID-19 mRNA vaccination and immune-related adverse events in patients with cancer receiving immune checkpoint inhibitors: a target trial emulation study.

Journal for immunotherapy of cancer·2026

Same author

Target Trial Emulation of Second-Generation Versus First-Generation Tyrosine Kinase Inhibitors in Adult Philadelphia Chromosome-Positive Acute Lymphoblastic Leukemia.

JCO oncology practice·2026

Same author

Predicting sleep state from continuous positive airway pressure flow in patients with obstructive sleep apnea.

Sleep medicine·2026

Same journal

Correction: Yang et al. Microstructural Characteristics of High-Pressure Die Casting with High Strength-Ductility Synergy Properties: A Review. <i>Materials</i> 2023, <i>16</i>, 1954.

Materials (Basel, Switzerland)·2026

Same journal

Effect of La and Ce Microalloying on the Corrosion Resistance of 0.4Sb Low-Alloy Steel in a Harsh Marine Atmospheric Environment.

Materials (Basel, Switzerland)·2026

Same journal

High-Temperature Properties of Magnesium Ammonium Phosphate Cement Modified with Gold Tailings.

Materials (Basel, Switzerland)·2026

Same journal

A Study on the Evolution of Intermetallic Phase Microstructure and High-Temperature Creep Behavior in Mg-8.0Al-1.0Nd-1.5Gd-Mn Alloys.

Materials (Basel, Switzerland)·2026

Same journal

Material-Driven Clinical Complications in Mechanical Circulatory Support: From Blood-Material Interactions to Device-Related Adverse Events.

Materials (Basel, Switzerland)·2026

Same journal

Influence of Final Irrigation on Calcium Silicate-Based Sealer Dentinal Tubular Penetration: A Systematic Review.

Materials (Basel, Switzerland)·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Aug 30, 2025

Three-Dimensional Particle Shape Analysis Using X-ray Computed Tomography: Experimental Procedure and Analysis Algorithms for Metal Powders

Three-Dimensional Particle Shape Analysis Using X-ray Computed Tomography: Experimental Procedure and Analysis Algorithms for Metal Powders

Published on: December 4, 2020

Deep Learning Applied to Defect Detection in Powder Spreading Process of Magnetic Material Additive Manufacturing.

Hsin-Yu Chen¹, Ching-Chih Lin¹, Ming-Huwi Horng²

¹Department of Mechanical Engineering, National Cheng Kung University, Tainan 701, Taiwan.

Materials (Basel, Switzerland)

|August 26, 2022

Summary

This summary is machine-generated.

This study introduces a convolutional neural network (CNN) for detecting defects in metal additive manufacturing (MAM) using selective laser melting (SLM). The developed method accurately identifies powder-spreading defects, improving product quality.

Keywords:

convolution neural network metal additive manufacturing powder-spreading defect selective laser melting

More Related Videos

Additive Manufacturing-Enabled Low-Cost Particle Detector

Additive Manufacturing-Enabled Low-Cost Particle Detector

Published on: March 24, 2023

Frequency Mixing Magnetic Detection Scanner for Imaging Magnetic Particles in Planar Samples

Frequency Mixing Magnetic Detection Scanner for Imaging Magnetic Particles in Planar Samples

Published on: June 9, 2016

Related Experiment Videos

Last Updated: Aug 30, 2025

Three-Dimensional Particle Shape Analysis Using X-ray Computed Tomography: Experimental Procedure and Analysis Algorithms for Metal Powders

Three-Dimensional Particle Shape Analysis Using X-ray Computed Tomography: Experimental Procedure and Analysis Algorithms for Metal Powders

Published on: December 4, 2020

Additive Manufacturing-Enabled Low-Cost Particle Detector

Additive Manufacturing-Enabled Low-Cost Particle Detector

Published on: March 24, 2023

Frequency Mixing Magnetic Detection Scanner for Imaging Magnetic Particles in Planar Samples

Frequency Mixing Magnetic Detection Scanner for Imaging Magnetic Particles in Planar Samples

Published on: June 9, 2016

Area of Science:

Materials Science
Manufacturing Engineering
Computer Vision

Background:

Metal additive manufacturing (MAM), particularly selective laser melting (SLM), offers high customization and rapid production for industries like medical, aerospace, and manufacturing.
Defects arising from thermal stress or hardware issues during SLM can compromise product quality, necessitating in-situ defect detection.
Powder-spreading defects, including uneven powder, uncovered areas, and recoater scratches, are common issues in SLM.

Purpose of the Study:

To develop an automated defect detection method for SLM processes using convolutional neural networks (CNNs).
To accurately identify and segment three types of powder-spreading defects: powder uneven, powder uncovered, and recoater scratches.
To evaluate the performance and computational efficiency of the proposed CNN model for real-time defect detection.

Main Methods:

Utilized process images captured by powder bed fusion equipment for defect detection.
Developed a two-stage CNN model: EfficientNet B7 for initial defect classification and instance segmentation networks for defect localization.
Evaluated Mask R-CNN with a ResNet 152 backbone for defect segmentation.

Main Results:

The Mask R-CNN model achieved high accuracy (0.9272) and Dice measurement (0.9438) in defect detection and segmentation.
The CNN model demonstrated efficient processing, with an approximate computational time of 0.2197 seconds per image.
The developed method effectively meets the requirements for early defect detection in SLM manufacturing.

Conclusions:

The proposed CNN-based approach provides an effective solution for automated defect detection in SLM.
The high accuracy and speed of the model contribute to improved quality control in metal additive manufacturing.
This method supports the reliable production of high-quality components in demanding industries.