大轴激光清洗的视觉检测方法研究  

Research on Visual Detection Method for Laser Cleaning of Large Shaft Parts

作　　者：陈偲 杨金堂[1] 周诗洋[1] 崔可涛 CHEN Cai;YANG Jin-tang;ZHOU Shi-yang;CUI Ke-tao(College of Mechanical and Automation,Wuhan University of Science and Technology,Hubei Wuhan 430081,China)

机构地区：[1]武汉科技大学机械自动化学院,湖北武汉430081

出　　处：《机械设计与制造》2023年第6期1-4,共4页Machinery Design & Manufacture

基　　金：国家重点专项资助项目(2018YFC1902400);国家自然科学基金(51805386)。

摘　　要：为实现大轴激光清洗效果的视觉检测判断,以大轴表面漆层为研究对象提出了一种清洗效果的判定方法,并对其可行性进行了实验分析。根据激光清洗工艺设计了自动清洗装置,使用工业相机采集漆层表面图像,通过中值滤波和直方图均衡对图像进行预处理,采用阈值图像分割法对图像进行特征提取,并设计了特征判断函数来完成对清洗效果的最终识别。实验结果表明,单次检测时间最大量级在0.1s,在高精度检测要求下,检测合格率能达到98%,符合大轴激光清洗检测在时间和准确性上的要求。In order to detect the laser cleaning effect of Large Shaft Parts with visual technology.The paint cleaning of Large Shaft Parts is taken as the research object,then a method to judge the cleaning effect is proposed,and the feasibility of this method is analyzed by experiments.An automatic cleaning device is designed according to the laser cleaning process,and the surface image of the paint layer is collected with industrial cameras.First,using Median filter algorithm and histogram equalization algorithm to achieve the purpose of preprocessing.Then,using threshold image segmentation algorithm to extract the image features.Finally,a feature judgment function is designed and using it to recognize the cleaning effect.The experiment shows that the maximum time level of each detection is 0.1s.Under the high precision requirements,the qualified rate of detection can also reach 98%,which can compliance with the detection requirements of time and accuracy.

关 键 词：机器视觉 激光清洗 目标检测 图像分割 

分 类 号：TH16[机械工程—机械制造及自动化]