基于区域分级的陶瓷阀芯表面缺陷检测系统研究与实现

发布时间：2018-01-10 19:09

本文关键词：基于区域分级的陶瓷阀芯表面缺陷检测系统研究与实现　出处：《组合机床与自动化加工技术》2017年10期 　论文类型：期刊论文

【摘要】：针对陶瓷阀芯表面缺陷特征复杂、啮合表面存在研磨抛光区域等问题,现有基于视觉检测的陶瓷阀芯表面缺陷检测算法对不同种类陶瓷阀芯的适应性较差且漏检率高,该研究提出了一种分区域、多级优化的陶瓷阀芯表面缺陷检测算法。首先依据陶瓷阀芯表面反射率的不同,将其划分为两个检测区域:抛光面(P)、非抛光面(Non P),然后根据陶瓷阀芯表面缺陷特征,将其分为两类缺陷:G缺陷、C缺陷。在检测过程中,搜索模板图像金字塔信息,调整目标灰度图像位姿,经掩模板(Mask)配准,提取出P区域、Non P区域,通过blob检测、裂纹检测,分别对P区域、Non P区域进行缺陷提取。通过实验研究与系统设计,分析了检测算法的可行性、适用性,结果表明该研究算法针对不同种类陶瓷阀芯表面缺陷检测漏检率低,且能通过优先级的合理设置,提高检测效率。
[Abstract]:Aiming at the complex surface defects of ceramic spool and the problem of grinding and polishing area on meshing surface, etc. The existing ceramic spool surface defect detection algorithm based on visual detection has poor adaptability and high leakage rate to different kinds of ceramic spool. Multi-level optimization of ceramic spool surface defect detection algorithm. Firstly, according to the different surface reflectivity of ceramic spool, it is divided into two detection areas: polished surface Pao, non-polished surface Non P). Then according to the characteristics of ceramic spool surface defects, it is divided into two types of defects: G defect and C defect. In the detection process, the template image pyramid information is searched, and the target gray image pose is adjusted. The P region was extracted by mask Mask registration, and the P region was detected by blob, crack detection, respectively. Non P region for defect extraction. Through experimental research and system design, the feasibility and applicability of the detection algorithm are analyzed. The results show that the detection rate of different ceramic spool surface defects is low, and the detection efficiency can be improved by setting the priority reasonably.
【作者单位】：湖北工业大学机械工程学院;
【基金】：湖北省教育厅重点项目:面向芯片贴装的显微视觉在线检测技术研究(D 20151406) 湖北省自然科学基金项目:多目视觉引导机器人绝对定位技术的研究(2016CFB513)
【分类号】：TH136;TP391.41
【正文快照】： 0引言陶瓷阀芯是由动片与静片组成的一种动密封组件,其因耐磨、耐高温、耐腐蚀、环保健康,广泛运用在各类水龙头中,通过控制进、出水孔间的偏移量调节水流大小。目前,随着国家强制性标准《陶瓷片密封水嘴》的实施,我国陶瓷阀芯需求量逐年增加,不同规模的生产企业也逐年增长。

【相似文献】