基于KNN分类的涂胶质量检测算法  

Gluing Quality Detection Algorithm Based on KNN Classification

作　　者：于一深 苏宇锋[1] 高建设[1] YU Yishen;SU Yufeng;GAO Jianshe(School of Mechanical and Power Engineering,Zhengzhou University,Zhengzhou 450000,China)

机构地区：[1]郑州大学机械与动力工程学院,郑州450000

出　　处：《组合机床与自动化加工技术》2023年第7期127-130,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(U1904169);国家重点研发计划项目(2018YFB0104100)。

摘　　要：涂胶作业是工业生产中常见的工艺流程,对密封性能有着极其重要的作用,它会直接影响密封件的质量,因此涂胶质量的缺陷检测方案的设计极其重要。为了能够快速高效地检测胶条是否符合要求,提出了一种基于KNN分类的涂胶质量检测算法。首先,通过坐标系转换和图像处理,获得涂胶轨迹的实际位置以及胶条宽度;其次,在轨迹中提取一定数量的采样点,获得采样点的胶条宽度和中心点偏移量,与标准宽度和标准轨迹进行比对便可以得到正确的检测结果。结果表明,该缺陷检测算法的准确率为97.85%,单个零件平均检测时间为98 ms,与传统人工检测相比,效率明显提高。The gluing operation is a common process in industrial production,which plays an extremely important role in the sealing performance.It will directly affect the quality of the seal.Therefore,the design of the defect detection scheme for the gluing quality is extremely important.In order to detect whether the glue strip meets the requirements quickly and efficiently,a glue quality detection algorithm based on KNN classification is proposed in this paper.First,The algorithm obtains the actual position of the gluing track and the width of the gluing strip through coordinate system transformation and image processing.After that,a certain number of sampling points are extracted from the trajectory,and the strip width and center point offset of the sampling points are obtained,and the correct detection result can be obtained by comparing with the standard width and standard trajectory.The detection results show that the accuracy of the defect detection algorithm is 97.85%,and the average detection time of a single part is 98 ms.Compared with the traditional manual detection,the efficiency is significantly improved.

关 键 词：机器视觉 涂胶质量检测 图像处理 KNN算法 

分 类 号：TH16[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]