基于机器视觉的波纹补偿器轴向尺寸检测

Axial Dimension Detection of Corrugated Compensator Based on Machine Vision

为了准确检测波纹补偿器的变形状况,改善波纹补偿器传统检测方式的弊端,提出了一种应用机器视觉的波纹补偿器轴向尺寸检测方法。首先,使用张正友相机标定法对工业CCD相机进行标定并采集波纹补偿器图像信息,然后提出一种改进Canny算子对图像进行边缘检测,获得可处理的理想化二值图像,遍历二值化图像进行水平倾斜校正后,提取极限边缘特征点。最后,计算出波纹补偿器在世界坐标系下的整体轴向尺寸和相邻波纹间的轴向尺寸。结果表明,该方法能较为精确的计算出波纹补偿器在不同变形程度下的轴向尺寸,并利用所计算出的尺寸数据准确地判断出了波纹补偿器的变形状况,对于保证波纹补偿器平稳工作,维护生产安全具有重要意义。

In order to accurately detect the deformation of the corrugated compensator and improve the disadvantages of the traditional detection method of the corrugated compensator,a detection method of the axial dimension of the corrugated compensator based on machine vision is proposed.ZHANG Zhengyou camera calibration method is used to calibrate the industrial CCD camera and collect the corrugated compensator image information,and then an improved Canny operator is proposed to detect the edge of the image to obtain a processable ideal binary image.After traversing the binary image for horizontal tilt correction,the limit edge feature points are extracted.Finally,the overall axial size of the ripple compensator in the world coordinate system and the axial size between adjacent corrugations are calculated.The experimental results show that the axial size of the corrugated compensator under different deformation degrees can be accurately calculated,and the deformation of the corrugated compensator can be accurately judged by the calculated size data,which is of great significance to ensure the smooth operation of the corrugated compensator and maintain production safety.