一种PCBA图像的快速定位方法  

The Fast Locate Method for PCBA Images

作　　者：李文龙 袁纵横[1] 孙伟 LI Wenlong;YUAN Zongheng;SUN Wei(School of Electronic Engineering and Automation,Guilin University of Electronic Technology,Guilin 541004,China;Anhui PinZhi Information Technology Co.Ltd,Hefei 230031,China)

机构地区：[1]桂林电子科技大学电子工程与自动化学院,广西桂林541004 [2]安徽品至信息科技有限公司,安徽合肥230031

出　　处：《电子科技》2018年第12期63-67,共5页Electronic Science and Technology

基　　金：广西自动检测技术与仪器重点实验室主任基金(NYQ14102)

摘　　要：针对实际项目中光学自动检测仪器高效率、高准确率的要求,单块较大PCBA板需要获取多张图像分别检测,根据图像特点寻找新的mark定位点而提出了一种改进的mark点定位方法。利用遍历法直接定位到最大概率的mark点坐标,并由先验数据计算出坐标点误差,根据可疑mark点坐标和坐标点误差代数和重新规划出新的兴趣区域,进而快速计算出准确的mark点坐标。计算结果表明,此方法能大幅减少冗余计算,新算法与原算法得到的坐标值完全相同,且新算法用时比原算法减少约40%。For the efficiency and high accuracy of the optical automatic testing instrument in the actual project, It was necessary to obtain multiple images separately for single block of PCBA, find a new mark location based on the image characteristics, and an improved mark point location method was proposed. Using the mark point coordinates which were directly located to the maximum probability by using ergotic calendar method, calculating the coordinate point enor by the prior data, according to the suspicious mark point coordinate and coordinate point algebra and re- programming the new interest area, and then quickly calculate the accurate mark point coordinates. The computation results showed that this method can reduce heavy redundant computation. The new algorithm＇s point coordinates was exactly the same with the original algorithm, and the calculate time reduces about 40% compared with the original al- gorithm.

关 键 词：印刷电路板 自动光学检测 图像识别 定位点 傅里叶变换 机器视觉 定位 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]