摘要
气门几何参数检测问题是汽车、船舶、航空以及柴油发动机等工业缺陷检测领域的热点难题,其本质是气门边缘关键点的检测问题,目前应用在工业气门产业的成果较少。本文针对气门边缘关键点难以精准检测问题,设计了基于亚像素边缘搜索算法,该算法基于Zernike矩和亚像素边缘检测对气门边缘进行矫正,并结合图像投影和逐像素边缘搜索精确定位边缘关键点对其拟合,得到气门几何参数检测结果。为了验证所提算法的有效性,对不同尺寸的工件进行检测分析,结果表明该算法检测气门杆径的最大绝对值误差和最大极差分别低于0.0093 mm和0.0363 mm,在检测精度和稳定性上均达到了工业要求,满足实际工业生产现场的检测需求。
Valve geometric parameter detection problem is a hot issue in industrial fields such as automobiles,ships,aviation,and diesel engines,which involves identifying critical points on valve edges.While progress has been made in related industrial valve applications,accurately detecting valve edge key points remains challenging.To address this difficulty,we design an edge search algorithm via sub-pixel.The algorithm corrects the valve edge based on Zernike moment and sub-pixel edge detection,and combines image projection and pixel-by-pixel edge search to accurately locate the key points of the edge,and fits them to yield valve geometric parameter detection outcomes.To verify the effectiveness of the proposed algorithm,the workpieces of different sizes are tested and analyzed.Experimental results demonstrate that maximum absolute error and the maximum extreme difference of the valve rod diameter detected by proposed algorithm are less than 0.0093 mm and 0.0363 mm,respectively.which meet the industrial requirements in terms of detection accuracy and stability,which can be applied to industrial production settings.
作者
汤华椿
谭棉
王前
罗太维
陈望
王林
TANG Huachun;TAN Mian;WANG Qian;LUO Taiwei;CHEN Wang;WANG Lin(School of Data Sciences and Information Engineering,Guizhou Minzu University,Guiyang 550025,China;Guizhou Key Laboratory of Pattern Recognition and Intelligent System,Guizhou Minzu University,Guiyang 550025,China)
出处
《智能计算机与应用》
2024年第1期49-55,共7页
Intelligent Computer and Applications
基金
贵州省科技计划项目(黔科合基础-ZK[2022]一般195,黔科合基础-ZK[2023]一般143,黔科合平台人才-ZCKJ[2021]007)
贵州省青年科技人才成长项目(黔教合KY字[2021]104,黔教合KY字[2021]111)
贵州省教育厅自然科学研究项目(黔教技[2023]061号,黔教技[2023]012号)。
