摘要
为了实现贴片机视觉系统对贴片元件的高精度定位,提出了一种基于Harris角点特征的贴片元件偏转角度与偏移量检测方法.首先对元件图像进行预处理,以改善引脚部分的图像质量,然后标记元件引脚的连通区域,判别元件引脚形状,再通过Harris角点特征提取、分类和直线拟合等得到元件的偏转角度,最后对元件图像进行旋转和放大处理,并进行基于Harris角点特征的图像配准,得到元件的偏移量.检测结果表明,文中算法能够检测贴片元件的偏转角度和偏移量,对偏转角度的检测误差小于0.1°,对偏移量的检测误差小于25μm,满足贴片机对视觉系统定位算法的精度要求.
In order to achieve high-precision positioning for chip components,a detection method was proposed based on Harris corner features to calculate deflection and offset.The component image was preprocessed to improve the quality of lead part,and the connected regions of component leads were marked to judge the lead shapes.The deflection was obtained by Harris corner features extraction,classification and straight line fitting.The component image was rotated and magnified,and two images were matched based on Harris corner features to obtain the offset.The results show that the proposed method can detect the deflection and offset with detection error less than 0.1° and 25 μm,which can meet the positioning accuracy requirement of placement machine vision system.
出处
《江苏大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2014年第5期531-535,共5页
Journal of Jiangsu University：Natural Science Edition
基金
国家自然科学基金资助项目(51175242)
江苏省科技成果转化专项资金资助项目(BA2012031)
江苏省普通高校研究生科研创新计划项目(CXZZ13_0436)
