Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components

Ma, Bei; Zhao, Jialong; Zhou, Shun; Wang, Hongjun; Xu, Junqi; Liu, Bingcai; Hou, Jingyao; Liu, Weiguo

doi:10.3390/mi16121373

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components

by

Bei Ma

,

Jialong Zhao

,

Shun Zhou

,

Hongjun Wang

,

Junqi Xu

,

Bingcai Liu

,

Jingyao Hou

and

Weiguo Liu

^*

School of Optoelectronic Engineering, Xi’an Technological University, Xi’an 710021, China

^*

Author to whom correspondence should be addressed.

Micromachines 2025, 16(12), 1373; https://doi.org/10.3390/mi16121373 (registering DOI)

Submission received: 10 September 2025 / Revised: 27 November 2025 / Accepted: 27 November 2025 / Published: 1 December 2025

(This article belongs to the Section A：Physics)

Download Versions Notes

Abstract

Optical components are extensively used in aerospace, microelectronic equipment, precision optical measurement, laser optics and other fields. Surface defects on optical components can significantly impact system performance, necessitating specialized detection methods. However, technical challenges persist in achieving high-resolution, high-precision and efficient optical surface defect detection. To address this, we propose an improved YOLOv8-based object recognition algorithm. By incorporating the BRA attention mechanism into YOLOv8’s backbone network, multi-scale feature maps are processed to enhance adaptability to complex scenarios. Simultaneously, replacing the feature fusion module with the Context-GuideFPN module enables contextual guidance and adaptive adjustments during multi-scale feature integration without excessive computational overhead. Experimental results on our high-quality microscopic dark-field image dataset demonstrate that the enhanced BACG-YOLOv8 achieves excellent performance in optical component defect detection. The optimized network accurately extracts defect details, particularly demonstrating refined edge feature extraction while effectively suppressing noise interference. This significantly reduces detection errors and improves defect extraction accuracy.

Keywords: optical components; surface defects; attention mechanism; deep learning

Share and Cite

MDPI and ACS Style

Ma, B.; Zhao, J.; Zhou, S.; Wang, H.; Xu, J.; Liu, B.; Hou, J.; Liu, W. Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components. Micromachines 2025, 16, 1373. https://doi.org/10.3390/mi16121373

AMA Style

Ma B, Zhao J, Zhou S, Wang H, Xu J, Liu B, Hou J, Liu W. Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components. Micromachines. 2025; 16(12):1373. https://doi.org/10.3390/mi16121373

Chicago/Turabian Style

Ma, Bei, Jialong Zhao, Shun Zhou, Hongjun Wang, Junqi Xu, Bingcai Liu, Jingyao Hou, and Weiguo Liu. 2025. "Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components" Micromachines 16, no. 12: 1373. https://doi.org/10.3390/mi16121373

APA Style

Ma, B., Zhao, J., Zhou, S., Wang, H., Xu, J., Liu, B., Hou, J., & Liu, W. (2025). Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components. Micromachines, 16(12), 1373. https://doi.org/10.3390/mi16121373

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Research on an Improved YOLOv8 Detection Method for Surface Defects of Optical Components

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI