Vision-based online detection system of support bar
Jun Chen, Chao Hu, Zaihe Yu, Jialin Cui, Xiaohe You · 2016
With the rapid development of modern manufacturing, more precision and higher efficiency are required for the detection technology. The vision-based detection system is playing a more important role in modern manufacturing because of its real-time and non-contact detection advantages. In this paper, we propose a vision-based online detection system for measuring diameter and thickness of the support bar which is a kind of auto parts. This system is based on industrial control computer platform and adopts back illuminating scheme. In this scheme, there are parallel backlight source to reduce interference of the support rod surface reflection and telecentric lens to decrease image distortion. We propose and compare the Line-based Shift Measurement Algorithm and the Point-based Measurement Algorithm, finally, select the latter with better anti-interference. Through the experiment, the maximum deviations of diameter and thickness are within 0.010mm and 0.007m. and the RMSE (Root Mean Square Error) is within 0.006mm.