摘要
针对汽车机械零部件中检测效率低、误检率高等瓶颈问题,提出了一种结合图像处理技术的方法。基于汽车副车架表面结构复杂以及缺陷异形等特征,首先通过小波邻域收缩去噪和多尺度增强算法对零件图像进行预处理,再通过水平集函数初步定位缺陷区域,最后结合空间模糊聚类进一步精确定位缺陷区域,达到对零件表面缺陷识别的目的。对工件表面划伤、压痕、气孔三种缺陷进行检测。实验结果表明:该算法相对于其它两种算法的检测率分别93.3%、86%和90%,具有较高的检测成功率。为后续缺陷零部件分类奠定良好的基础且可以满足工业检测要求。
A method the combined with image processing technology is proposed,to meet the bottleneck problems of low detection efficiency and high error rate in automobile mechanical parts.It is due to the characteristics of complex surface structure and different defects shape to automobile subframe.Firstly,the method of the wavelet neighborhood shrinking and denoising is used to pretreated the surface image of the vehicle subframe.Secondly,the defect area is located roughly by based on the level set function.Subsequently,the surface defect recognition will be defected further accurately by the spatial fuzzy C-means clustering.The surface scratch,indentation and air hole of the workpiece were detected.The experimental results show that the detection rate of this algorithm is 93.3%,86%and 90%respectively compared with the other two algorithms,and it has a high detection success rate.The purpose is to lay a good foundation for subsequent defect classification and meet the requirements of industrial inspection.
作者
林晶
问轲
张学昌
刘永跃
LIN Jing;WEN Ke;ZHANG Xue-chang;LIU Yong-yue(Harbin University of Commerce,Heilongjiang Harbin 150028,China;Ningbo Institute of Technology of Zhejiang University,Zhejiang Ningbo 315100,China;Ningbo Heli Mould Technology Co.,Ltd.,Zhejiang Ningbo 315700,China)
出处
《机械设计与制造》
北大核心
2024年第7期317-321,325,共6页
Machinery Design & Manufacture
基金
宁波市科技创新2025重大专项项目的资助(2019B10099)
黑龙江省属高校科技成果研发项目(TSTAU-R2018009)。
关键词
铝合金铸件
缺陷检测
图像处理
水平集
Aluminum Alloy Casting
Defect Detection
Image Processing
Level Set
