基于Hough变换的高精度机器视觉对准系统的研究

发布时间：2018-03-26 18:05

本文选题：对准系统　切入点：霍夫变换　出处：《应用光学》2017年06期

【摘要】：随着集成电路特征尺寸的不断减小,I/O引脚数日益增加。对于高密度封装,尤其是2.5D/3D封装中,电子元件和基底快速而准确的对准十分关键。利用高性能CCD搭建了一套精密的光学视觉对准系统,并采用Hough变换算法对芯片和基底进行对准试验。采集的图像经过去噪预处理后采用形态学边缘检测方法提取对齐标记,并通过Hough变换得到4组平行直线,然后计算基底的位移和旋转角度并完成对准。使用Matlab编程配准,程序运行时间约为4.2s,旋转参数的误差小于1.2°,x和y轴的平移误差均小于1pixel。试验结果表明基于Hough变换算法的光学视觉对准系统可以快速而精确地实现芯片和基底对准,满足IC封装需求。
[Abstract]:With the decreasing of IC feature size, the number of I / O pins is increasing. For high-density packaging, especially in 2.5D/3D packaging, Fast and accurate alignment of electronic components and substrates is very important. A set of precise optical vision alignment system is built by using high performance CCD. Hough transform algorithm is used to test the alignment of chip and substrate. After de-noising preprocessing, the image is extracted by morphological edge detection method, and four groups of parallel lines are obtained by Hough transform. Then calculate the displacement and rotation angle of the substrate and finish the alignment. Use Matlab programming registration, The running time of the program is about 4.2 s, the error of rotation parameter is less than 1.2 掳x and the translation error of y axis is less than 1 pixel. The experimental results show that the optical vision alignment system based on Hough transform algorithm can realize chip and substrate alignment quickly and accurately. Meet IC packaging requirements.
【作者单位】：唐山学院机械工程省级实验教学示范中心;江苏师范大学机电工程学院;华中科技大学数字制造装备与技术国家重点实验室;
【基金】：国家自然科学基金(51305179) 数字制造与装备技术国家重点实验室开放基金(DMETKF2016005)
【分类号】：TN40;TP391.41

【相似文献】