焊缝裂纹缺陷筛选及识别技术研究

Research on Screening and Identification Technology for Weld Crack Defects

磁粉检测是碳钢原件表面和近表面最有效的无损检测手段之一,但在复杂工况下存在缺陷识别率低、出现误报等情况。本研究基于磁粉检测自动化技术,开展裂纹缺陷筛选及识别技术研究,使用Otsu自适应阈值算法对裂纹缺陷图像进行分割,提取目标缺陷图像,分析噪声、伪缺陷和裂纹图像的特征参数,通过特征筛选,保留真实裂纹信息,并对裂纹图像进行标记处理,完成裂纹数量和长度的识别。

Magnetic particle testing is one of the most effective non-destructive testing methods for the surface and near surface of carbon steel components.In complex working conditions,there are still low defect recognition rates and false alarms.This study is based on the automation technology of magnetic particle testing,and conducts research on crack defect screening and recognition technology.The Otsu adaptive threshold algorithm is used to segment crack defect images,extract target defect images,analyze noise,pseudo defects,and crack image feature parameters.Through feature selection,real crack information is retained,and crack images are marked and processed to complete the identification of crack quantity and length.