基于深度学习的废旧电路板表面缺陷分类
Deep learning-based surface defect classification of scrap circuit boards
余光海 1盛光鸣 1张元昊 1徐观明2
作者信息
- 1. 合肥工业大学机械工程学院,合肥 230009
- 2. 湖南绿色再生资源有限公司,长沙 410699
- 折叠
摘要
针对废旧电路板表面缺陷识别,提出了一种利用深度学习进行分类的方法.在ResNet34模型的基础上,增加了CBAM注意力机制.通过将ResNet34架构与注意力机制相结合,旨在增强模型关注输入图像中重要特征并对其进行优先排序的能力,从而提高分类的准确性.结果表明提出的方法在提高废旧电路板表面缺陷分类方面的有效性.
Abstract
A classification method using deep learning is proposed for the recognition of surface defects on scrap circuit boards.Based on the ResNet34 model,the CBAM attention mechanism is added.By combining the ResNet34 architecture with the attention mechanism,it aims to enhance the model's ability to focus on important features in the input image and prioritize them,thus improving the classification accuracy.The results demonstrate the effectiveness of the proposed method in improving the clas-sification of surface defects on scrap circuit boards.
关键词
废旧电路板/注意力机制/表面缺陷分类Key words
waste printed circuit boards/attention mechanism/surface defect classification引用本文复制引用
基金项目
国家重点研发计划重点专项项目(2019YFC1908002)
出版年
2024
NETL
NSTL
万方数据