焊缝视觉图像处理和跟踪控制系统地研究.pdf

Thesis Submitted to the A cademic mittee, Hebei University of Technology for The Master Degree of Control Science and Engineering A RESEARCH ON SEAM IMAGE PROCESSING AND VISUAL TRACKING CONTROL SYSTEM by LI Jiangwei Supervisor : Professor Li Lianbing December 2011 河北工业大学硕士学位论文 i 焊缝视觉图像处理与跟踪控制系统的研究摘要在对钢板进行焊接时,因采用“示教”模式而越来越不能满足复杂多变的工业生产的要求。具有视觉感知能力和较高智能性的基于结构光焊缝焊后质量检测与跟踪技术,有助于提高系统的自动化水平,提高劳动生产率。在本文中,根据不同焊缝自身的形状特点以及现场实际的情况,选择合适的摄像机、光源以及镜头等,建立合适的几何位置关系。视觉系统的硬件结构搭配好以后,对摄像机采集的视频设计并组合图像预处理算法。根据图像预处理完成后的图像特点,提取线结构光几何中心条纹,设计焊缝信息识别与提取算法,同时兼顾算法的实时性与鲁棒性。基于视觉系统的宏微机器人,焊前示教以及焊时通过视觉系统实时矫正相结合的方式,提高了焊机的性能。宏微机器人分为宏动部分和微动部分,通过视觉采集的几点建立的焊缝数学模型,宏动部分会沿着根据数学模型示教好的轨迹运动,同时微动部分根据视觉系统反馈回来的焊缝信息进行实时矫正运动。这样保证了在不同种类(如直线焊缝、曲线焊缝等) 的焊缝中,能够精确的进行焊接,同时对焊缝质量进行检测。本文采用 LabVIEW 软件平台,相比于其他文本类编程语言,其图形化的编程环境和强大的子函数库使得工程师可较快实现工程目标,函数库中具有一些视觉图像处理算法, 其运算速度以及鲁棒性也是其他编程语言所无法比拟的,进一步保证了检测的实时性。利用 LabVIEW SQL Toolkit 工具包建立基于 Access 的数据库系统,可以实现相关数据的快速储存、查看、删除和其他相关处理。关键词: 图像处理,跟踪控制,宏微机器人, LabVIEW 焊缝视觉图像处理与跟踪控制系统的研究 ii A RESEARCH ON SEAM IMAGE PROCESSING AND VISUAL TRACKING CONTROL SYSTEM ABSTRACT During laser welding, it cannot adapt to the ever changing working environment for using “teach” mode. The weld quality detection systems based on structured light machine vision is a key tool which can help to improve the overall level of automation and productivity. In the thesis, we selected the right camera, li ght source and lens, and built their relationship of their geometric position according to different shapes and the actual situation. We had to design bine the image pre-processing algorith m after the hardware structure of the visual system had a good future. And then, the geometri c center line stripes of structured light are extracted. The algorithm which is real-time an d robust had been designed for identifying and extracting seam information. Based on the visual sense, the robot is capable of effectively realizing the teaching mode a nd real-time tracking mode si multaneo