基于Gamma校正HSV模型的变色镀铜管脚检测  

Detection of copper-plated pin discoloration based on HSV model with Gamma correction

作　　者：李研 车艳秋[1,2] 梅健强 LI Yan;CHE Yanqiu;MEI Jianqiang(School of Automation and Electrical Engineering,Tianjin University of Technology and Education,Tianjin 300222,China;Tianjin Noise Hazard Engineering Protection Technical Guidance Center,Tianjin 300222,China;School of Electronic Engineering,Tianjin University of Technology and Education,Tianjin 300222,China)

机构地区：[1]天津职业技术师范大学自动化与电气工程学院,天津300222 [2]天津市噪声危害工程防护技术指导中心,天津300222 [3]天津职业技术师范大学电子工程学院,天津300222

出　　处：《天津职业技术师范大学学报》2022年第3期1-6,I0002,共7页Journal of Tianjin University of Technology and Education

基　　金：天津市教委科研计划项目(JWK1605).

摘　　要：针对芯片封装基板上镀铜管脚变色缺陷检测存在的人工目视检测效率低、误检率高、传统视觉检测系统装置复杂以及检测方法性能低等问题,提出了一种新的特征提取方法。该方法将芯片封装基板图像从RGB(红-绿-蓝)颜色空间转换至HSV(色调-饱和度-明度)颜色空间,在HSV颜色阈值法预处理阶段引入Gamma校正,再进行平滑去噪和形态学处理。研究测试了不同色彩空间模型对变色镀铜管脚的检测能力,结果表明,Gamma校正HSV颜色模型的检测精度达到98.11%,高于普通HSV颜色模型(95.46%)和RGB颜色模型(10.10%)的检测精度,为机器视觉在变色镀铜管脚快速检测的应用提供了依据。A new feature extraction method is proposed to address the problems of low artificial visual detection efficiency,high false detection rate,complex traditional visual detection system and device,and low performance of the detection method for copper-plated pins discoloration on chip packaging substrates.Using this method,the image was converted from RGB(red-green-blue) to HSV(hue-saturation-lightness) color space,then Gamma correction was introduced in the HSV color thresholding preprocessing stage,and finally smoothing denoising and morphological processing were performed.The detection ability of different color space models for discolored copper-plated pins was tested.The results show that the detection accuracy of the Gamma-corrected HSV color model reaches 98.11 %,which is higher than that of the ordinary HSV color model(95.46%) and the RGB color model(10.10%).It provides a basis for the application of machine vision in the rapid detection of discolored copper-plated pins.

关 键 词：HSV颜色模型 变色镀铜管脚 GAMMA校正 形态学处理 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TQ153[自动化与计算机技术—计算机科学与技术]