摘要
提出了一种基于Halcon的钢球表面缺陷检测系统设计方案。首先,利用CCD相机采集展开盘上的钢球表面图像,并传输到应用软件;其次,进行图像感兴趣区域选取、滤波去噪和缺陷边缘检测等图像处理;然后,将缺陷区域进行连通,并计算缺陷面积;最后,与缺陷阈值进行比较,控制分球机械手与分球板协同作用实现分拣。试验结果表明:该系统具有很高的检测效率和准确率,可满足工业自动化生产的需求,具有良好的应用前景。
A design scheme of detecting system for surface defects on steel balls is proposed based on Halcon. Firstly,the CCD camera is used to capture surface image of steel balls,and then the image is transmitted to application software. Secondly,the image processing is carried out,such as selecting region of interest,filtering,denoising and detecting defect edge,and then the defect regions are connected and calculated. Finally,comparing with defect threshold,the synergistic effect between ball splitting manipulator and ball splitting plate is controlled to realize sorting. The experimental result shows that the system features high detecting efficiency and accuracy. The requirements are met for industrial automatic production,and the good application prospect is provided.
引文
[1]何加群.中国工业强国战略和轴承产业[J].轴承,2015(1):55-63.
[2] LIN Y Z,LIU X L,HAN H Y,et al. Detection and recognition of steel ball surface defect based on MATLAB[J].Key Engineering Materials,2009,416:603-608.
[3]潘卫清,赵晓波.钢球表面缺陷的双波长干涉数字相位检测[J].中国激光,2014,41(5):205-210.
[4]栗琳,王仲,裴芳莹,等.双图像传感器的球表面展开方法[J].仪器仪表学报,2012,33(7):1641-1646.
[5] LI L,WANG Z,PEI F Y,et al. Improved illumination for vision-based defect inspection of highly reflective metal surface[J]. Chinese Optics Letters,2013,11(2):35-38.
[6]栗琳.基于视觉的高反射球面缺陷快速检测关键技术研究[D].天津:天津大学,2013.
[7]王义文,屈冠彤,刘献礼,等.钢球表面缺陷的图像差分检测算法[J].计算机辅助设计与图形学学报,2016,28(10):1699-1704.
[8] WANG Y W,PEI Z C,LIU X L,et al. Based on machine vision encoder grating assembled eccentric adjustment system design[J]. Application Research of Computers,2013,30(12):3814-3816.
[9]王成林.钢球表面缺陷的多光纤检测方法研究[D].济南:济南大学,2014.
[10] WANG C L,AI C S,LI G P. Optical fiber sensor system for detecting the steel ball surface quality[J]. Applied Mechanics&Materials,2014,654:245-249.
[11] ZHAO Y L,CHE W B,ZHOU K,et al. Design of deployment mechanim of steel ball and surface defects recognition[J]. Applied Mechanics&Materials,2013,274:245-248.
[12]刘青,张晓晖,黄军勤.基于机器视觉的钢球表面缺陷快速提取算法[J].激光杂志,2014(9):58-61.
[13]王文成.基于Halcon的齿轮缺陷检测系统设计[J].机械传动,2014(9):60-63.
[14] ZHAO C L,DAI M,XIONG J Y. Region-of-interest based rate control for UAV video coding[J]. Optoelectronics Letters,2016,12(3):216-220.
[15]王建新,朱恩,刘文松,等.基于JPEG2000的遥感图像感兴趣区域编码新算法及其VLSI设计[J].电子与信息学报,2016,38(4):958-963.
[16]黄文柯,张肖宁.利用形态学多尺度算法分割粗集料粘连图像[J].哈尔滨工业大学学报,2016,48(3):125-130.
[17]施隆照,王凯.基于连通区域的复杂车牌的字符分割算法[J].计算机工程与设计,2016,37(8):2138-2142.
