华南理工大学学报(自然科学版) ›› 2013, Vol. 41 ›› Issue (3): 116-121,128.doi: 10.3969/j.issn.1000-565X.2013.03.017

• 机械工程 • 上一篇    下一篇

基于颜色特征的智能焊点定位算法

谢宏威1 张坤2 邝泳聪2 张宪民2 裴海龙1   

  1. 1.华南理工大学 自动化科学与工程学院,广东 广州 510640;2.华南理工大学 广东省精密装备与制造技术重点实验室,广东 广州 510640
  • 收稿日期:2012-10-23 修回日期:2012-11-14 出版日期:2013-03-25 发布日期:2013-02-01
  • 通信作者: 谢宏威(1981-),男,博士后. 主要从事视觉检测、模式识别研究. E-mail:xhw_cn@foxmail.com
  • 作者简介:谢宏威(1981-),男,博士后. 主要从事视觉检测、模式识别研究.
  • 基金资助:

    广东省科技计划项目( 2012B011300016) ; 粤港关键领域重点突破项目( 东莞专项2012205122)

Intelligent Positioning Algorithm of Solder Joints Based on Color Features

Xie Hong-wei1 Zhang Kun2 Kuang Yong-cong2 Zhang Xian-min2 Pei Hai-long1   

  1. 1.School of Automation Science and Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China;2.Guangdong Provincial Key Laboratory of Precision Equipment and Manufacturing Technology,South China University of Technology,Guangzhou 510640,Guangdong,China
  • Received:2012-10-23 Revised:2012-11-14 Online:2013-03-25 Published:2013-02-01
  • Contact: 谢宏威(1981-),男,博士后. 主要从事视觉检测、模式识别研究. E-mail:xhw_cn@foxmail.com
  • About author:谢宏威(1981-),男,博士后. 主要从事视觉检测、模式识别研究.
  • Supported by:

    广东省科技计划项目( 2012B011300016) ; 粤港关键领域重点突破项目( 东莞专项2012205122)

摘要: 针对现有自动光学检测系统焊点定位自动化程度低的问题,提出了一种基于颜色特征的智能焊点定位算法.对样本焊点区域的灰度直方图进行统计分析,获得焊点分割的初始颜色阈值,并以最小像素损失为目标对阈值进行修正,得到最优分割阈值,实现了焊点颜色的自动抽取; 定位过程中,使用阈值对焊点图像进行二值化,并使用优化后的面积最大法来搜索焊点位置,实现了焊点的快速定位. 实验结果表明: 文中算法对焊点图像的分割效果理想,定位误差小,定位速度可以满足实际应用的要求.

关键词: 焊点定位, 自动光学检测, 阈值, 面积最大法, 二值化

Abstract:

In order to improve the automation degree of solder joint positioning by using the existing automatic opticalinspection systems,an intelligent positioning algorithm based on color features is proposed.In this algorithm,theinitial color thresholds of solar joint segmentation are calculated by statistically analyzing the gray histograms of solderjoint regions,and a minimum pixel loss method is used to correct the initial thresholds and to further obtain theoptimal threshold.Thus,the solder joint color is automatically extracted.Moreover,during the positioning,the imagesof solder joints are binarized with the obtained threshold,and the optical location of solder joints are fastsearched by using the optimized maximum area method.Experimental results show that the proposed algorithm iseffective because it helps to obtain properly-extracted solder joint images with small positioning error and high positioningspeed that meets the practical application requirements well.

Key words: solder joint positioning, automatic optical inspection, threshold, maximum area method, binarization

中图分类号: