| 操乐文.基于改进YOLOv5的冲压件缺陷检测方法研究[J].安徽建筑大学学报,2024,32(): |
| 基于改进YOLOv5的冲压件缺陷检测方法研究 |
| Research on Defect Detection Method of Stamping Parts Based on Improved YOLOv5 |
| 投稿时间：2022-11-25 修订日期：2022-12-29 |
| DOI： |
| 中文关键词: 冲压件 缺陷检测 注意力机制 |
| 英文关键词: stamping defect detection attention mechanism |
| 基金项目:基于图论的无监督多比特量化编码算法研究，国家自然科学基金（青年）（62001004） |
|
| 摘要点击次数: 649 |
| 全文下载次数: 174 |
| 中文摘要: |
| 摘要：冲压件在生产的过程中容易出现裂纹、划痕、起皱、凹凸点等缺陷问题，而目前生产线上对于冲压件的缺陷检测还是以人工检测为主，不仅效率低还容易造成漏检。本文针对以上问题，提出了一种基于改进YOLOv5模型的缺陷检测算法。为了更好的聚焦缺陷，在YOLOv5模型中引入CA注意力模块。通过对比实验，将目标框损失函数改为GIoU，进一步提升定位精度。实验表明相较于原模型，改进后的YOLOv5模型精准度，召回率，mAP值均得到提升。 |
| 英文摘要: |
| Abstract: Stamping parts are prone to defects such as cracks, scratches, wrinkles, and bumps in the production process. At present, the defect detection of stamping parts on the production line is mainly manual detection, which is not only inefficient but also prone to missed inspections. . Aiming at the above problems, this paper proposes a defect detection algorithm based on the improved YOLOv5 model. In order to better focus on the defects, a CA attention module is introduced into the YOLOv5 model. Through comparative experiments, the loss function of the target frame is changed to GIoU to further improve the positioning accuracy. Experiments show that compared with the original model, the improved YOLOv5 model has improved accuracy, recall rate and mAP value. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
