Related Experiment Video
Updated: Aug 8, 2025
08:35
Luminescence Lifetime Imaging of O2 with a Frequency-Domain-Based Camera System
Published on: December 16, 2019
9.3K
MO-SOD: Micro-Oxidation Small Object Detection Model for Oxygen-Free Copper Surfaces Based on Microscopic Imaging
Qianqian Li1,2, Taohong Zhang1,2, Mingyang Yang3
1Department of Computer, School of Computer and Communication Engineering, University of Science and Technology Beijing (USTB), Beijing 100083, China.
ACS Omega
|February 27, 2023
Summary
Detecting micro-oxidation in oxygen-free copper is challenging. A new MO-SOD model uses AI for rapid, accurate detection, improving quality control in materials science.
Area of Science:
- Materials Science
- Artificial Intelligence
- Computer Vision
Background:
- Micro-oxidation is a critical defect in precision oxygen-free copper, difficult to detect visually.
- Manual microscopic inspection is slow, costly, and subjective.
Purpose of the Study:
- To develop an automated system for rapid and accurate micro-oxidation detection on oxygen-free copper surfaces.
- To introduce the Micro-Oxidation Small Object Detection (MO-SOD) model for real-time defect analysis.
Main Methods:
- Developed the MO-SOD model integrating small target feature extraction, attention pyramid, and an anchor-free detector.
- Enhanced the model with a combined CIOU and focal loss function for improved detection accuracy.
- Trained and validated the model on a dataset of oxygen-free copper surface micrographs with varying oxidation levels.
Main Results:
- The MO-SOD model achieved a mean Average Precision (mAP) of 82.96% in detecting micro-oxidation spots.
- Demonstrated superior performance compared to existing advanced object detection models.
- Successfully integrated with a high-definition microphotography system for robotic platform application.
Conclusions:
- The MO-SOD model offers a highly efficient and accurate solution for automated micro-oxidation detection.
- This technology can significantly improve quality control processes for oxygen-free copper materials.
- The proposed method paves the way for real-time, on-site defect identification in industrial settings.
