基于灰度信息的PCB板精确定位  

Grayscale-Information-based Image Locating of PCBs

作　　者：崔怀峰[1] 杜遥雪[1] 

机构地区：[1]五邑大学机电工程学院,广东江门529020

出　　处：《五邑大学学报（自然科学版）》2010年第4期59-64,共6页Journal of Wuyi University(Natural Science Edition)

基　　金：广东省自然科学基金资助项目(9151063101000021)

摘　　要：利用模板匹配法在待测PCB图像中建立坐标系,采用基于边缘灰度幅度的轮辐法检测定位圆,由幅度排序相关搜索算法检测出最佳圆;提出PCB板角度定位的3种算法,比较了不同算法在检测精度、速度和稳定性上的差异,实证了轮辐法可实现PCB板的精确快速定位.Image locating of PCBs is largely influenced by the automatic detection of PCB defections.Pattern matching techniques are used to establish a coordinate system for images of the PCB.Then a spoke method was used to find points on a circular edge and find the optical circle.Three algorithms for the angle orientation of the detected PCB image were given and compared.Theoretical analysis and experiments show the spoke measure is efficient.

关 键 词：PCB板 定位圆 模板匹配 轮辐法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]