摘要
目的:针对目前汽车仪表盘质量检测中采用人工读取指针示数方法导致的效率低下、误差较大等问题提出一种基于ZS细化改进算法的汽车仪表盘指针读数方法.方法:首先利用面阵CCD相机获取汽车仪表盘原始图像并进行预处理,运用ZS细化算法获取初始细化图像;然后采用先斜线再十字的搜索和方向判别方法对ZS细化算法进行改进,实现了对初始细化图像的充分细化;最后利用最小二乘法进行直线拟合,并计算出指针读数.结果:实验表明,该方法解决了ZS快速并行细化算法中局部像素冗余、细化直线存在分叉的问题,从而提高了直线识别的准确度.结论:该方法较原算法的平均相对误差降低了0.55%.
Aims:Aiming at the problems of the manual reading of pointers in the quality inspection of automobile dashboards,a method of the pointers reading based on ZS refinement improved algorithm is proposed.Methods:First,the area CCD camera is used to obtain the original image of the car dashboard and pre processed.The ZS thinning algorithm is used to obtain the initial thinned image.Then,the search and direction discrimination method of the oblique line and cross first is used to improve the ZS thinning algorithm.In order to fully refine the initial thinning image,the least square method was used to perform a straight line fitting,and the pointer reading was calculated.Results:The experimental results show that the proposed method solves the problem of local pixel redundancy and refinement of the straight line in the ZS fast parallel refinement algorithm,and improves the accuracy of line recognition.Conclusions:Compared with the original algorithm,the average relative error of this method is reduced by 0.55%.
作者
齐淑雯
陈爱军
刘磊
QI Shuwen;CHEN Aijun;LIU Lei(College of Metrology and Testing Engineering,China Jiliang University,Hangzhou 310018,China)
出处
《中国计量大学学报》
2020年第1期44-49,共6页
Journal of China University of Metrology
