Abstract: Compared with traditional manual detection methods of defects in vials, online machine vision inspection method has tremendous superiority. Therefore an automation inspection system based of machine vision for vials is presented. Firstly, the mechanical and electrical structure of the system is studied, and the visual imaging and image acquisition devices are designed. Secondly, according to the particularity of each parts of vials, the paper puts forward specific detection methods. Finally, it selects 100 bottles of various types of defects for testing, the results show that the system has a high detection accuracy and can satisfy the requirement of the actual production.

Key words: vial defects, visual inspection, center location, circle scan

摘要： 与传统的西林瓶缺陷人工检测方法相比，在线视觉自动检测具有巨大优越性。设计了一套西林瓶缺陷自动视觉在线检测系统，研究了系统的机械与电气结构，并设计了针对西林瓶的视觉成像与图像获取装置；根据西林瓶各部位的形状特殊性，针对性地提出了具体的检测方法。选取各类缺陷样品100瓶进行测试，结果表明该系统检测准确度较高，能满足实际生产的需求。

关键词: 西林瓶缺陷, 视觉检测, 圆心定位, 圆周扫描