基于机器视觉技术的螺纹识别系统 thread recognition system based on machine vision technology.pdfVIP

第28 卷第5 期 光电技术应用 Vol.28 ，No.5 2013 年10月 ELECTRO-OPTIC TECHNOLOGY APPLICATION October ，2013 ·光电系统· 基于机器视觉技术的螺纹识别系统 景 敏 （陕西理工学院机械工程学院，陕西 汉中 723003） 摘 要：螺纹牙型角识别是区分螺纹种类的常用手段，传统检测手段效率低、量规易磨损、成本高，已不能满足现代工业高效 发展的需求。利用CCD 获取螺纹基本图像，并通过图像的平滑、边缘检测、二值化处理及轮廓提取，对螺纹轮廓进行分析，从中 测量出螺纹的牙型角参数并进行识别。探讨了利用机器视觉对螺纹牙型角参数进行测量的方法，并设计出一套以机器视觉识别 技术为核心的视觉传感和图像处理系统为一体的螺纹识别系统。从理论和实践上证实了该方法的可行性和准确性。 关键词：螺纹牙型角；机器视觉；边缘检测；轮廓提取；识别系统 中图分类号：TH89 文献标识码：A 文章编号：1673-1255（2013 ）-05-0006-04 ThreadRecognitionSystemBasedonMachineVisionTechnology JING Min (School of Mechanical Engineering, Shanxi University of Technology, Hanzhong 723003, China) Abstract:Thread angle identification is a common method to distinguish thread types. Traditional detection methods have many disadvantages such as low efficiency and high cost and gauges are easy to be abraded. The needs of high efficient development of modern industry are not met any more. CCD is used to obtain the basic im⁃ age of thread. And the thread contour is analyzed through image smoothness, edge detection, binary image pro⁃ duction and contour hunting. The thread angle parameters are measured and identified. The measurement meth⁃ ods of thread angle parameter using machine vision are discussed. And a thread recognition system mainly based on the machine vision recognition technology and integrated visual sensing with image processing system is de⁃ signed. The feasibility and correctness of the method is proved from theory and practice. Keywords:thread angle; machine vision; edge detection; contour hunting; recognition system 螺纹连接应用于机械行业中的各个领域。常见 数测量法，使用专用的量规进行接触测量以获得单 螺纹种类有普通螺纹、管螺纹、梯形螺纹等。螺纹牙 个参数。二者都需人工操作，设备体积大，检测程序 型角的区别是区分螺纹种类的常用手段，普通螺纹 多，整个测量过程耗时耗力，工作效率低，检测成本 牙型角为60° ，管螺纹