基于轴线局部搜索的曲轴轴径视觉测量方法

Visual measurement method of crankshaft diameter based on axis local search

采用传统的机器视觉测量方法对曲轴的几何特征进行测量时,存在测量精度偏低的问题,为此,提出了一种基于轴线局部搜索的曲轴轴径视觉测量方法(测量系统)。首先,依据待测零件所需的要求,进行了硬件选型和机器视觉测量平台的搭建,校正了镜头畸变,提高了其精度;对所获取的图像进行了预处理,包括形态学滤波处理、二值化处理等操作;其次,通过定位算法像素级坐标,获取了曲轴旋转轴线位置,使其与相机移动轴线相重合;运用亚像素细分算法提取了亚像素边缘坐标,计算了其像素当量,完成了像素与实际尺寸之间的转换,实现了对轴径进行高精度测量的目的;最后,进行了测量实验的对比,对基于轴线局部搜索的轴径测量方法的准确性和实用性进行了验证。研究结果表明:轴线局部搜索法测量轴径平均误差为0.003 mm,且测量的稳定性较好,证明了该算法及其测量系统的准确性和实用性,可以适用于对曲轴轴径进行有效检测。

When the traditional machine vision measurement method is used to measure the geometric characteristics of crankshaft,the measurement accuracy is low.Therefore,a visual measurement method(measurement system)of crankshaft diameter based on local search of axis was proposed.Firstly,according to the requirements of the parts to be tested,the hardware selection and machine vision measurement platform were built.The lens distortion was corrected to improve the precision,and the acquired image was preprocessed,including morphological filtering,binarization and other operations.Then the position of the crankshaft rotation axis was obtained by pixel level coordinates of the positioning algorithm to make it coincide with the camera moving axis.The subpixel subdivision algorithm was used to extract the subpixel edge coordinates,calculate the pixel equivalent to complete the conversion between the pixel and the actual size,and achieve high-precision measurement of the shaft diameter.Finally,the accuracy and practicability of the shaft diameter measurement method based on local search of the axis were verified by comparing the measurement experiments.The research results show that the average error of the axis local search method for measuring the shaft diameter is 0.003 mm,and the stability is good,which proves the accuracy and practicability of the algorithm and its measurement system.It can be used to effectively detect the crankshaft diameter.