基于OpenCV的工件亚像素内径测量研究

基于OpenCV的工件亚像素内径测量研究(中文4800字,英文1600字)
摘要
本文提出了一种基于OpenCV的非接触测量方法，实现了工件的亚像素内径测量。 设计并实施内径可视测量系统测量工件的过程。 在图像中，噪声和目标的灰度非常接近，以至于很难被正确的阈值分开。 对于处理这个难点，提出了两个步骤中获取两个不同阈值的方法。首先，用较小的阈值去除背景中的噪声; 第二步，使用更大的阈值使二值化图像接近目标的实际轮廓。 系统处理图像进行二值化，对所需区域进行截取，轮廓提取等过程，最终实现工件的高精度内径测量。 在视野中不同位置重复测量工件多次，实验结果表明该方法的一致性优于HALCON机器视觉软件。
关键字：计算机视觉；轮廓检测；子像素测量；
Research of sub-pixel inner diameter Measurement of workpiece Based on OpenCV
Abstract—A method of untouched measurement  based on OpenCV is proposed to achieve the sub-pixel inner diameter Measurement of workpiece. Design and implement a visual measurement system on the inner diameter Measurement of workpiece. In the image, the grayscale of the noise and the aim is very close and very hard to be separated by the right threshold. To this difficult point, a method of two thresholds in two steps is proposed. First, remove the background noise with a smaller threshold; second, get the binarization image close to the real contour of the aim with a bigger threshold. The system deals with the image binarization, the interception of region of interest, contour extraction and so on, finally achieves the high-precision inner diameter measurement of workpiece. Repeat the measurement many times of the workpiece in different place in the field of view, the experimental results show that the consistency of the method is better than the HALCON machine vision software.
Eeywords: OpenCV; computer vision; contour detection; sub-pixel measurement;