基于深度学习的木板材表面缺陷识别算法研究
Research on Wood Surface Defect Recognition Algorithm Based on Deep Learning
沈锦桃 1王祺 1李欢 1孔维亮 1钱绍祥2
作者信息
- 1. 镇江中福马机械有限公司,江苏镇江 212127
- 2. 镇江高等专科学校,江苏镇江 212310
- 折叠
摘要
我国是木材及木制品的生产、制造和出口大国,木板材表面缺陷会降低木板材的外在品质和内部强度,进而影响木材的加工生产过程.基于自制的木材表面缺陷数据集,在YOLOv5s模型上针对时下主流的几种损失函数进行了综合测试与对比分析.旨在拓展深度学习模型在木材缺陷检测领域的应用,以期为木材表面缺陷自动化检测提供新思路.
Abstract
China is a large country in the production,manufacturing and export of wood and wood products,and the surface defects of wood panels will reduce the external quality and internal strength of wood panels,which will affect the processing and production process of wood.Based on the self-made wood surface defect data set,this paper con-ducts comprehensive testing and comparative analysis of several mainstream loss functions on the YOLOv5s model.This paper aims to expand the application of deep learning model in the field of wood defect detection,in order to provide a new idea for automatic inspection of wood surface defects.
关键词
木板材/缺陷识别/损失函数/YOLOv5s/自动化检测Key words
wood plank/defect identification/loss function/YOLOv5s/automatic detect引用本文复制引用
出版年
2024
NETL
NSTL
万方数据