2022年第46卷第10期76器件与应用artsandApplicationsP文献引用格式:李颖,姜红,肖莉莉,等.基于Blob分析的手机屏幕缺陷检测方法[J].电声技术,2022,46(10):76-79.LIY,JIANGH,XIAOLL,etal.MobilephonescreendefectdetectionmethodbasedonBlobanalysis[J].AudioEngineering,2022,46(10):76-79.中图分类号:TP751文献标识码:ADOI:10.16311/j.audioe.2022.10.022基于Blob分析的手机屏幕缺陷检测方法李颖,姜红,肖莉莉,王兆森,李春辉(信阳学院教育学院,河南信阳464000)摘要:针对手机屏幕缺陷检测的高效率、高精度的要求,提出一种基于Blob分析的手机屏幕缺陷检测方法。首先使用基于PatMax算法的模板定位技术对图像进行定位,其次使用基于仿射变换的图像校正技术实现模板图像与测试图像的逐像素对齐,最后采用Blob分析算法获取缺陷特征的位置、面积、方向等信息。采用文中所提出的方法、差影法对200幅手机屏幕图像进行缺陷检测,实验结果显示,所提方法的准确率为98%,误检率为1.5%,漏检率为0.5%,每张图像的平均检测时间为52.19ms。相较于差影法,所提出的方法具有更高的准确率和检测速度。关键词:Blob分析;手机屏幕;缺陷检测;差影法MobilePhoneScreenDefectDetectionMethodBasedonBlobAnalysisLIYing,JIANGHong,XIAOLili,WANGZhaosen,LIChunhui(CollegeofEducation,XinyangUniversity,Xinyang464000,China)Abstract:ThispaperproposesadefectdetectionmethodformobilephonescreenbasedonBlobanalysistomeetthehighefficiencyandhighprecisionrequirementsofmobilephonescreendefects.Firstly,thetemplatelocalizationtechnologybasedonPatMaxalgorithmisusedtolocatetheimage,andtheimage-correctiontechniquebasedonaffinetransformationisusedtorealizethepixel-by-pixelalignmentofthetemplateimageandthetestimage.Finally,theBlobanalysisalgorithmisusedtoobtainthelocation,area,directionandotherinformationofdefectfeatures.Thedefectsof200mobilephonescreenimagesweredetectedbytheproposedmethodandthedifferenceimagemethod.Theexperimentalresultsshowthattheaccuracyoftheproposedmethodis98%,thefalsedetectionrateis1.5%,andthemisseddetectionrateis0.5%.Theaveragedetectiontimeofeachimageis52.19ms.Comparedwiththedifferenceimagemethod,theproposedmethodhashigheraccuracyanddetectionspeed....