期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2019
卷号：2239
页码：67-72
出版社：Newswood and International Association of Engineers
摘要：Thread inspection is a very important step in
modern industrial inspection and is also the key to guaranteeing
product quality. However, the traditional contact measurement
method is time-consuming and expensive. This study developed
an automatic noncontact measurement system that can be used
to on-machine measure thread dimension based on machine
vision for computer numerical control (CNC) lathe machine
tools. During the measurement process, backlighting was used to
obtain the image of a thread profile. Then, a machine vision
algorithm was developed to analyze the obtained thread images
in order to capture the thread profile coordinate points and
compute the dimensions of the thread pitch and thread depth.
The experimental results showed that the maximum difference
between the measured dimensions and real dimensions is around
6 µm (1 pixel resolution), thus verifying the feasibility and
accuracy of the on-machine external thread measurement
system developed in this study.
