基于机器视觉的SOP芯片引脚缺陷检测系统设计

SOP Chip Pins Defect Detection Based on Machine Vision System Design

为解决人工检测SOP芯片引脚缺陷所存在的诸多问题,设计一套基于机器视觉的SOP芯片引脚缺陷自动检测系统。系统采用ARM-DSP双核架构快速地实现图像采集、处理和特征提取以及外围应用功能,设计基于动态阈值的快速分割算法来获取芯片的特征图像,采用灰度跃变检测引脚中点、中点直线拟合和引脚间距统计等方法实现对SOP芯片引脚缺陷的自动检测。实验结果表明该系统快速、准确、可行。

In order to solve the problems in manual inspection of SOP chip defect, design an automatic detection system of SOP chip defect based on machine vision. The system uses ARM-DSP dual-core architecture to achieve these functions rapidly such as image acquisition, processing and feature extraction as well as peripheral applications. It designed a fast segmentation algorithm based on dynamic threshold to get the chip feature image and used the methods of grayscale transition detector pin midpoint, midpoint line fitting and pin spacing statistical to realize automatic detection of the SOP chip pin defect. Experimental results show that the system is fast, accurate and feasible.