双列深沟球轴承滚动体缺失检测技术研究

发布时间：2018-06-09 14:18

本文选题：机器视觉 + 虚拟仪器　；参考：《河南科技大学》2012年硕士论文

【摘要】：在双列深沟球轴承装配过程中，滚动体缺失是最常见的质量问题。国内各轴承厂对滚动体主要都依靠人工方式来检测。这种原始的人工检测方式有很多不可靠的因素，如工人的情绪、视觉疲劳等，容易发生漏检和误检，无法有效保证轴承产品的合格率。机器视觉技术可以弥补人工检测的不足，不仅大大节省人力，而且也可提高检测的速度、准确度，实现生产与检测同步自动化，为轴承企业带来可观的经济效益。目前，针对双列滚动体的视觉检测技术研究还处在起步阶段。本文提出了一种基于机器视觉的双列深沟球轴承滚动体缺失检测系统，用于检测双列深沟球轴承的滚动体。双列滚动体在用传统方式采集图像时，由于表面光滑和相互干扰，缺失特征不明显，不能用以往的视觉系统进行检测，因此本系统采用了新型的硬件平台和检测软件。硬件平台包含上位机和下位机两部分，上位机由PC计算机构成，负责轴承图像的处理和识别，下位机以单片机为核心，设计了专用的控制电路，负责控制分拣轴承次品等执行机构。为了实现双列轴承的自动翻转，提高检测的效率，还设计了翻转机构，当系统检测完一列滚动体，自动翻转双列轴承后检测另一列滚动体，从而实现了对整个双列深沟球轴承滚动体的快速检测。软件部分采用了新型的算法，利用多个同心圆来实现轴承中待检区域的精确定位；然后，通过逐环扫描展开法，提取滚动区域，，并利用面积统计滚动体的数目。最后，在所提取的区域中用标准块扫描，对缺失的位置进行定位，再通过坐标变换进行显示标记，从而完成检测，相比以往的机器视觉产品，检测更准确，性能更加可靠。本系统在双列深沟球轴承装配线上进行了现场测试，可以有效地完成双列滚动体检测，该系统具有运行稳定，快速高效、抗干扰强等优点，达到系统设计的目标的要求。
[Abstract]:In the assembly process of double row deep groove ball bearings, the loss of rolling body is the most common quality problem. The shaft bearing plants in China mainly rely on manual detection of the rolling body. There are many unreliable factors in this original manual testing method, such as worker's emotion, visual fatigue and so on. It is easy to miss and misdetect, which can not effectively guarantee the qualified rate of bearing products. Machine vision technology can make up for the shortage of manual inspection. It not only saves manpower greatly, but also improves the speed and accuracy of detection, realizes the automation of production and detection synchronously, and brings considerable economic benefits to bearing enterprises. At present, the research of vision detection technology for double-row rolling body is still in its infancy. This paper presents a new system based on machine vision for detecting the scrolling body of double-row deep groove ball bearing, which can be used to detect the rolling body of double-row deep groove ball bearing. Because of the smooth surface and mutual interference, the missing feature is not obvious, so it can not be detected by the previous visual system. Therefore, the system adopts a new hardware platform and testing software. The hardware platform consists of two parts: the upper computer and the lower computer. The upper computer is composed of PC computer, which is responsible for the processing and recognition of bearing image. Designed a special control circuit, responsible for the control of sorting bearing defects and other executive agencies. In order to realize the automatic turnover of double-row bearings and improve the efficiency of detection, a turning mechanism is also designed. When the system detects one row of rollers, it automatically flips the double-row bearings and detects another row of rollers. In the software part, a new algorithm is used to realize the accurate location of the area to be detected in the bearing by using multiple concentric circles, and then the ring-by-ring scanning expansion method is used to realize the accurate location of the rolling body of the double-row deep groove ball bearing. The rolling area is extracted and the number of rolling objects is counted by the area. Finally, using standard block scanning in the extracted area, locating the missing position, then displaying and marking by coordinate transformation, the detection is completed. Compared with the previous machine vision products, the detection is more accurate. The system has been tested on the assembly line of double-row deep groove ball bearings, and can effectively complete the double-row rolling body detection. The system has the advantages of stable operation, fast and high efficiency, strong anti-interference and so on. Meet the requirements of system design.
【学位授予单位】：河南科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH133.3

【参考文献】