基于机器视觉的气门油封装配质量在线检测

The Real-Time Detection of Valve Seal Assembly Quality Based on Machine Vision

目前在汽车发动机装配线上仍主要依靠人工视觉检测气门油封装配质量,效率低且易出错。引入机器视觉技术,利用智能相机获取压装后的气门油封图像,应用图像定位、模板匹配、特征提取等算法,实现对气门油封装配质量的在线自动检测。实践证明:该方法可替代人工对气门油封压装质量进行100%在线检测,快速稳定,满足生产要求。

At present ,in the assembly line of automobile engine is still mainly depends on manual visual inspection of the assembly quality of valve stem oil seal, which always accompany with low efficiency and high error ratio. Introducing the machine vision technology, which includes using intelligent camera to obtain the valve stem seal image after the press-fit, combining with image positioning algorithm, template matching algorithm and feature extraction algorithm, to realize the automatic detection of the assembly quality of the valve stem oil seal. Practice shows that this method can replace the manual visual inspection of valve stem oil seal, can absolutely ensure the assembling quality of online detection, fast and stable, and also can meet the production requirements.