摘要
生产的精整圆坯表面会出现烧伤、接柱、针孔、夹渣、刮伤、渣沟、缩孔等缺陷。由于圆坯较长,表面无法通过单次测量获取,所以传统的精整圆坯表面缺陷检测是由若干名工人目视进行的,其工作效率与检测精度已无法满足现代化智能生产的需要。设计了一套基于人工智能的多目视觉精整圆坯滚检表面缺陷检测系统。从多目视觉系统采集的图像数据中精准地检测出缺陷,并依据系统标定结果确定缺陷位置,完成缺陷的快速准确检测。现场生产工况表明,该系统能显著提升既有设备的生产效率与自动化水平,有助于实现精整圆坯生产流程的降本增效与智能运作。
The surface of the production of finished round billet produced is prone to burn,column connection,pinhole,slag inclusion,scratch,slag groove,shrinkage and other defects.The traditional method to identify the defect of the finished round billet is carried out by the visual inspection of several workers.Its work efficiency and detection accuracy can not meet the needs of modern intelligent production.In this paper,a set of artificial intelligence-based multi-eye vision finishing round billet rolling inspection surface defect detection system is designed.Defects are accurately detected from the image data collected by the multi-eye vision system,and the defect location is determined according to the system calibration results to complete the rapid and accurate detection of defects.
出处
《工业控制计算机》
2023年第8期58-60,共3页
Industrial Control Computer
关键词
精整圆坯
缺陷检测
机器学习
多目视觉
finished round billet
defect detection
machine learning
multi-eye vision
