Wood Panel Defect Detection Based on Improved YOLOv8n

Wood panel surface defect detection is critical to product quality. Traditional detection methods are time-consuming and subjective, and they can lead to economic waste, while deep learning image recognition techniques offer a new approach. However, the accuracy and convergence speed of existing def...

Full description

Saved in:
Bibliographic Details
Main Authors: Rui Li, Shilu Zhong, Xuemei Yang
Format: Article
Language:English
Published: North Carolina State University 2025-02-01
Series:BioResources
Subjects:
Online Access:https://ojs.bioresources.com/index.php/BRJ/article/view/24141
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849736504409063424
author Rui Li
Shilu Zhong
Xuemei Yang
author_facet Rui Li
Shilu Zhong
Xuemei Yang
author_sort Rui Li
collection DOAJ
description Wood panel surface defect detection is critical to product quality. Traditional detection methods are time-consuming and subjective, and they can lead to economic waste, while deep learning image recognition techniques offer a new approach. However, the accuracy and convergence speed of existing defect detection techniques still require improvement. In this paper, an improved algorithm based on YOLOv8n was designed for accurate detection of wood panel defects. The C-ADown method was designed to replace traditional downsampling, while preserving high-frequency features. The combination of the Dilation-wise Residual Module and multi-scale dilation attention was employed to enhance the multiscale robustness of defect detection. A hybrid encoder was added to improve localization accuracy. The loss function was optimized to improve detection accuracy and convergence speed. Compared to the base YOLOv8 version, the improved model achieved a 6.1% increase in mAP, an 8% increase in recall, and a 3.6% increase in precision, significantly enhancing the model’s detection capabilities. The GitHub link to the improved algorithm files is as follows: (https://github.com/humblefactos1/YOLOV8-CDC/tree/main.)
format Article
id doaj-art-ec9553d796c043e8935e4419d1824f35
institution DOAJ
issn 1930-2126
language English
publishDate 2025-02-01
publisher North Carolina State University
record_format Article
series BioResources
spelling doaj-art-ec9553d796c043e8935e4419d1824f352025-08-20T03:07:16ZengNorth Carolina State UniversityBioResources1930-21262025-02-01202255625732384Wood Panel Defect Detection Based on Improved YOLOv8nRui Li0Shilu Zhong1Xuemei Yang2College of Furnishings and Industrial Design, Nanjing Forestry University, Nanjing 210037, ChinaCollege of Furnishings and Industrial Design, Nanjing Forestry University, Nanjing 210037, ChinaCollege of Furnishings and Industrial Design, Nanjing Forestry University, Nanjing 210037, ChinaWood panel surface defect detection is critical to product quality. Traditional detection methods are time-consuming and subjective, and they can lead to economic waste, while deep learning image recognition techniques offer a new approach. However, the accuracy and convergence speed of existing defect detection techniques still require improvement. In this paper, an improved algorithm based on YOLOv8n was designed for accurate detection of wood panel defects. The C-ADown method was designed to replace traditional downsampling, while preserving high-frequency features. The combination of the Dilation-wise Residual Module and multi-scale dilation attention was employed to enhance the multiscale robustness of defect detection. A hybrid encoder was added to improve localization accuracy. The loss function was optimized to improve detection accuracy and convergence speed. Compared to the base YOLOv8 version, the improved model achieved a 6.1% increase in mAP, an 8% increase in recall, and a 3.6% increase in precision, significantly enhancing the model’s detection capabilities. The GitHub link to the improved algorithm files is as follows: (https://github.com/humblefactos1/YOLOV8-CDC/tree/main.)https://ojs.bioresources.com/index.php/BRJ/article/view/24141wood panel deep learningyolov8nc-adowndilation-wise residualmulti-scale dilation attentionloss function
spellingShingle Rui Li
Shilu Zhong
Xuemei Yang
Wood Panel Defect Detection Based on Improved YOLOv8n
BioResources
wood panel
deep learning
yolov8n
c-adown
dilation-wise residual
multi-scale dilation attention
loss function
title Wood Panel Defect Detection Based on Improved YOLOv8n
title_full Wood Panel Defect Detection Based on Improved YOLOv8n
title_fullStr Wood Panel Defect Detection Based on Improved YOLOv8n
title_full_unstemmed Wood Panel Defect Detection Based on Improved YOLOv8n
title_short Wood Panel Defect Detection Based on Improved YOLOv8n
title_sort wood panel defect detection based on improved yolov8n
topic wood panel
deep learning
yolov8n
c-adown
dilation-wise residual
multi-scale dilation attention
loss function
url https://ojs.bioresources.com/index.php/BRJ/article/view/24141
work_keys_str_mv AT ruili woodpaneldefectdetectionbasedonimprovedyolov8n
AT shiluzhong woodpaneldefectdetectionbasedonimprovedyolov8n
AT xuemeiyang woodpaneldefectdetectionbasedonimprovedyolov8n