摘要
红外辐射成像检测由于其非接触性、快速性、检测面积大等特点,成为当前主要的无损检测方法。基于传热学理论知识建立了热量在钢板中传导的数学模型,并利用红外无损检测技术对含有缺陷的钢板进行研究。通过求取图像矩阵方差判断缺陷,采用直方图和形态学方法对缺陷图片进行预处理,提取缺陷的几何特征。使用阈值分割方法将缺陷从背景中分离出来,达到了图像的平滑滤波、缺陷中心定位、边缘检测等目的。提出了计算缺陷面积和周长的算法,完成对缺陷面积和周长等几何特征的检测。该研究方法及结论对钢板红外无损检测具有一定的指导意义。
This paper based on the knowledge of heat transfer theory to establish a mathematical model of heat conduction in the steel plate and use the technology of infrared nondestructive testing to study on steel plate with defects.Judging defects by Striking matrix variance of image and using histogram and morphology to pretreatment the defective picture and extracting geometric characteristics of defects.using the method of threshold segmentation to separate defects from background,reached the purposes of smoothing filter,defect center position detection and edge detection.Proposed an algorithm to calculate area and perimeter of the defect,complete the detection of the defect geometric features such as area and perimeter.
出处
《工业控制计算机》
2012年第12期54-56,共3页
Industrial Control Computer
关键词
无损检测
热传导
图像处理
缺陷识别
non-destructive testing,thermal conductivity,image processing,defect recognition
