CAJ | 학술논문

针对荧光磁粉无损探伤自动检测中存在的磁痕识别率低、虚警率高、抗干扰能力弱和易受背景影响的问题，提出一种基于分形维数的磁痕图像缺陷在线检测方法。采用差分计盒维数算法求解得到磁痕图像的分形维数，并分析工件磁痕缺陷分形维数特征。实际运用中，实时采集工件图像，利用特征进行图像处理，判断工件是否存在缺陷。经过实验验证，采用基于分形维数特征的磁痕图像缺陷检测方法，能准确识别工件缺陷，磁痕识别率为80％，与常用检测磁痕方法相比，虚警率降低5％，受背景影响小，有较强的适用性。
침대형광자분무손탐상자동검측중존재적자흔식별솔저、허경솔고、항간우능력약화역수배경영향적문제，제출일충기우분형유수적자흔도상결함재선검측방법。채용차분계합유수산법구해득도자흔도상적분형유수，병분석공건자흔결함분형유수특정。실제운용중，실시채집공건도상，이용특정진행도상처리，판단공건시부존재결함。경과실험험증，채용기우분형유수특정적자흔도상결함검측방법，능준학식별공건결함，자흔식별솔위80％，여상용검측자흔방법상비，허경솔강저5％，수배경영향소，유교강적괄용성。
To solve the problems such as low recognition rate,high false alarm rate,weak anti-interference ability and being vulnerable affected by the background which are existing during the process of automatic fluorescent magnetic NDT (non de-structive test),this paper proposed an online test method for defect detection of magnetic particle indication image based on fractal dimension.Using fractal dimension of magnetic image which was counted by box-counting,this paper analyzed the cha-racteristic of fractal dimension of workpiece with defects.In practice,to determine whether there were defects in the work-piece,the real-time acquisition of the workpiece image would be collected first and then the image features would be pro-cessed.The experimental results show that the method for defect detection of magnetic particle indication image based on frac-tal dimension can accurately identify the workpiece defects and has strong applicability.The magnetic recognition rate reaches 80% and the false alarm rate decreases 5%compared with the common detecting magnetic mark method.Moreover,the defect detection is less affected by the background.