当前位置：首页 > 过刊浏览->2023年44卷第10期
熔模精密铸件荧光自动检测技术研究进展及 智能化发展趋势
Research Progress and Intelligent Development Trend of Fluorescent Automatic Detection for Investment Castings
浏览(2706) 下载(6)
- DOI:
- 作者:
- 余慧澎 1 ，康茂东 1 ， 2 ，王 俊 1 ， 2
YU Huipeng 1 , KANG Maodong 1,2 , WANG Jun 1,2
- 作者单位:
- 1. 上海交通大学 材料科学与工程学院，上海 200240；2. 上海市先进高温材料及其精密成形重点实验室，上海 200240
1. School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; 2. Shanghai Key Lab of Advanced High-Temperature Materials and Precision Forming, Shanghai 200240, China
- 关键词:
- 机器视觉；机器学习；熔模铸造；荧光检测；表面缺陷
machine vision; machine learning; investment casting; fluorescent penetrant inspection; surface defect
- 摘要:
- 熔模精密铸件表面缺陷严重降低铸件的服役可靠性。 工程上 , 铸件表面缺陷的检测通常采用荧光检测方法。 然而，由于荧光检测的图像复杂、检测员水平参差不齐和长时间检测引起的视觉疲劳等原因，降低了荧光检测的精度和效率，严重影响航空航天重大装备服役安全。 近年来，荧光自动检测技术逐渐发展起来。 本文系统总结了国内外荧光自动检测系统的研究现状，给出了基于传统处理法建立荧光自动检测系统的主要步骤，梳理了近年来新出现的基于深度学习法的荧光自动检测方法，并对未来荧光缺陷智能化检测的发展趋势进行了展望。The surface defects of investment castings seriously reduce the reliability of castings in service. In engineering, the surface defects of castings are usually detected by fluorescent penetrant inspection (FPI). However, due to the complexity of the image, the uneven level of inspectors and the visual fatigue caused by long-term inspection, the accuracy and efficiency of FPI are reduced. Therefore, fluorescent penetrant automatic inspection systems have been gradually developed. This paper systematically summarizes the research status of fluorescence penetrant automated inspection at home and abroad. This paper also gives the main steps of the automatic fluorescent defect detection system based on the traditional image processing method, and reports a new automatic fluorescent defect detection module based on the deep learning method. Furthermore, this paper predicts the development trend of intelligent fluorescent defect detection in investment castings.