基于Harris-Hough算法的芯片初始测试点对准方法  

Method on alignment of chip initial test point based on Harris-Hough algorithm

作　　者：梅迪 苏中[1] 刘洪[1] MEI Di, SU Zhong, LIU Hong(Beijing Key Laboratory of High Dynamic Navigation Technology, Beijing Information Science ＆ Technology University, Beijing 100101, China)

机构地区：[1]北京信息科技大学高动态导航技术北京市重点实验室,北京100101

出　　处：《中国测试》2018年第9期121-125,共5页China Measurement & Test

基　　金：北京市科技计划课题(Z161100003016011);高动态导航技术北京市重点实验室开放课题(HDN2017002)

摘　　要：针对芯片测试设备中的机器视觉系统可以实现芯片自动对准的需求,提出一种基于Harris-Hough算法的芯片初始测试点对准方法。通过Harris角点检测算法获得图像中探针顶端的角点位置,建立图像位置定位基准;通过Hough变换提取芯片初始测试点位置,计算芯片初始测试点和探针顶端之间所需移动距离;在不同光学放大倍率下进行粗、细两步对准,达到所需定位精度要求。该方法经过Matlab编程实现,经测试验证,能够有效获得芯片初始测试点及探针顶端的位置坐标,准确度达到0.3μm。经过距离换算,可以为芯片测试设备的运动控制系统提供移动参数。A method on automatic alignment of chip initial test point based on Harris-Hough algorithm is proposed aiming at the requirement of automatic alignment of chips on the chip testing equipment. A image positioning benchmark is established, when Harris corner detection algorithm is used to obtain the location of the probe tip. The location of the chip initial test point is obtained by Hough transform and the distance between the initial test point and probe tip is calculated. Under different optical magnifications, coarse alignment and fine alignment are carried out to get the requirement of positioning accuracy. The method is implemented by Matlab programming. After testing, it can effectively obtain the position coordinates between the chip initial test point and probe tip and the accuracy is up to 0.3 μm. Through the distance conversion, it can provide moving parameters for the motion control system of the chip testing equipment.

关 键 词：芯片测试 HARRIS角点检测 HOUGH变换 自动对准 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]