Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Scan Path Optimization and YOLO-Based Detection for Defect Inspection of Curved and Glossy Surfaces.

Sensors (Basel, Switzerland)·2026

Same author

A Physics-Guided Dual-Sensor Framework for Bearing Fault Diagnosis in PMDC Motor Drives.

Sensors (Basel, Switzerland)·2026

Same author

Optimization of Flow Rate for Uniform Zinc Phosphate Coating on Steel Cylinders: A Study on Coating Uniformity and Elemental Composition Using Scanning Electron Microscopy (SEM).

Materials (Basel, Switzerland)·2025

Same author

Hyperelastic and Stacked Ensemble-Driven Predictive Modeling of PEMFC Gaskets Under Thermal and Chemical Aging.

Materials (Basel, Switzerland)·2024

Same author

LSTM-Autoencoder for Vibration Anomaly Detection in Vertical Carousel Storage and Retrieval System (VCSRS).

Sensors (Basel, Switzerland)·2023

Same author

Correlation Coefficient Based Optimal Vibration Sensor Placement and Number.

Sensors (Basel, Switzerland)·2022

Search research articles

相关实验视频

Updated: May 10, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

使用深度学习和先进的成像系统,优化光泽和曲面上的缺陷检测.

Joung-Hwan Yoon¹, Chibuzo Nwabufo Okwuosa¹, Nnamdi Chukwunweike Aronwora¹

¹Department of Mechanical Engineering (Department of Aeronautics, Mechanical and Electronic Convergence Engineering), Kumoh National Institute of Technology, 61 Daehak-ro, Gumi-si 39177, Gyeonsangbuk-do, Republic of Korea.

Sensors (Basel, Switzerland)

|April 26, 2025

概括

这项研究引入了一种改进的AI方法,用于检测光泽,曲的产品上的缺陷. 定制的CNN模型为工业应用提供了高精度和卓越的计算效率.

关键词:

狄克斯特拉的算法是什么?这就是ResNet-50的特点.在VGG-16中.卷积神经网络是一种卷积神经网络.曲的表面表面的曲线.错误分类错误分类错误分类错误分类检测故障的检测故障检测.一个光亮的表面表面.

更多相关视频

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Published on: August 18, 2022

相关实验视频

Last Updated: May 10, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Published on: August 18, 2022

科学领域:

计算机视觉计算机视觉
人工智能的人工智能
工业自动化工业自动化

背景情况:

人工智能 (AI) 越来越多地被工业界采用,因为它的效率.
人工智能支持的图像分析用于缺陷检测和质量控制.
传统的人工智能方法由于反射性而难以在光泽和曲的表面上检测缺陷.

研究的目的:

开发一种增强的人工智能方法,以改进图像数据收集,以便在具有挑战性的表面上检测缺陷.
训练和评估深度学习模型,以准确有效地检测故障.
确定用于工业部署的计算强大和高效的AI模型.

主要方法:

使用了带有专门照明的巴斯勒视觉摄像头和KEYENCE移位传感器来增强图像采集.
训练了八个深度学习算法,包括自定义的卷积神经网络 (CNN),VGG-16和ResNet-50变体.
采用正常和两个缺陷条件的图像数据进行模型培训和评估.

主要成果:

ResNet-50₂₂₄实现了最高的精度 (97.97%) 的损失为0.1030.
CNN₆-240表现出卓越的计算效率,平均步骤时间为94毫秒,准确率为95.08%.
该研究确定了测试模型中的整体准确性和计算效率之间的权衡.

结论:

增强的数据收集方法改善了AI在光泽,曲的表面上的缺陷检测.
ResNet-50₂₂₄提供高精度,而CNN₆-240则适用于资源有限的环境.
这些发现为选择适合工业缺陷检测的AI模型提供了宝贵的见解.