基于机器学习的铝熔体夹渣自动检测技术
Automatic Detection Technology for Inclusion in Al Melt Based on Machine Learning
白蕊 1胡勇 1金泽发 2刘宏泉 3闫志杰4
作者信息
- 1. 太原科技大学材料科学与工程学院,太原 030024
- 2. 宁夏大学材料与新能源学院,银川 750021
- 3. 栋梁铝业有限公司,湖州 313000
- 4. 中北大学材料科学与工程学院,太原 030051;特殊环境先进金属材料山西省重点实验室,太原 030051
- 折叠
摘要
根据现有夹渣图像的特点,提出基于YOLOv5模型的夹渣目标检测算法,以减少获取图像的角度、光源等不确定因素对测试结果所造成的负反馈影响,提高检测精度.利用Mosaic数据增强、自适应锚框计算、自适应图像缩放等技术,融合Focus和CSP结构,设计出基于YOLOv5的自动化识别夹渣图像和自动计算夹渣率的优化算法.结果表明,相对于人工采集照片计算夹渣率水平的方法,改进后的YOLOv5s模型,有效提高了断面夹渣图像目标检测的精确度,由改进前的83%提高至97%.
Abstract
An optimized target detection algorithm based on YOLOv5 model was proposed according to characteristics of existing slag inclusion images to reduce the negative feedback effects caused by uncertain factors such as angle and light source,improving the accuracy.According to Mosaic data enhancement,adaptive anchor frame calculation,adap-tive image scaling and other technologies,an algorithm for automatic recognition of slag inclusion images and auto-matic calculation of slag inclusion rate was designed combined with Focus and CSP structures.The results indicate that the improved YOLOv5s model can effectively enhance the target detection accuracy of sectional slag inclusion images from 83%to 97%,compared with the manually collecting method to calculate slag inclusion rate level.
关键词
熔体质量检测/夹渣/YOLOv5/目标检测Key words
Melt Quality Inspection/Slag Inclusion/YOLOv5/Target Detection引用本文复制引用
基金项目
山西省自然科学基金资助项目(202103021224279)
山西省科技成果转化引导专项资助项目(202204021301025)
山西省科技创新人才团队资助项目(202204051002020)
出版年
2024
NETL
NSTL
万方数据