面向回转类零件圆度的机器视觉测量方法与试验

Method and Experimental Study of Machine Vision Measurement for The Roundness of Rotary Parts

针对工程应用中回转类零件的圆度测量精度不高、效率低等问题,提出了基于改进的Zernike矩亚像素边缘特征提取算法的回转类零件圆度视觉测量方法。建立了回转类零件轴截面图像的机器视觉测量系统,在分析零件轴截面图像特征的基础上,将传统的Zernike矩算法与最大类间方差法相结合得到改进的快速算法,提高了亚像素边缘点的定位精度。建立了轴截面圆度测量数学模型,应用标准量块对系统进行标定,实现了回转类零件直径与圆度的非接触式测量。试验结果表明,该方法不仅可使直径与圆度的测量达到亚像素级精度,还避免了传统测量方法造成的低效率与误判,实现了快速、准确、非接触式测量。

Aiming at the problem of low accuracy and low efficiency of the roundness error measurement of the rotary parts in engineering application, the method of measuring the roundness error of the rotary parts based on the improved Zernike moment sub pixel edge feature extraction algorithm is proposed. The machine vision measurement system based on the image of the axis of rotary parts is established. Based on the analysis of the image features of the parts, the traditional Zernike moment algorithm and the maximum class variance method are improved. The shaft cross - section roundness measurement mathematical model is established the system is calibrated by the standard gauge the non-contact measurement of the roundness error of revolving parts. The experimental results indicate that the measured rotational parts of roundness error is 0.03μm and the roundness error of the cone is 0.01 p,m, and the measuring precision is improved greatly.