微型机与应用Issue(20)：32-34,3.
基于荧光磁粉图像的缺陷检测技术
Defect detecting technology based on fluorescent magnetic image
初延亮 1肖宇峰 1刘桂华 1张华1
作者信息
- 1. 西南科技大学 信息工程学院 特殊环境机器人技术四川省重点实验室，四川 绵阳 621010
- 折叠
摘要
Abstract
Fluorescent magnetic nondestructive examination is widely used for its advantages of visual display of defect , high sensitivity, fast detection and low cost. Based on the analyses of display characteristics of the fluorescent magnetic image, smoothing and de-noising algorith and segmentation algorithm are studied. It propose simage smoothing algorithm based on weighted template and adaptive neighborhood selection, and applies algorithm of Ridler adaptive threshold to segmentation of defects image. Experimental results show that this method can remove the non-defects and extract the real defects of work piece from the background of image completely, which lays a good foundation for automated fluorescent magnetic nondestructive examination.关键词
荧光磁粉检测/加权模板/自适应邻域选择/Ridler 自适应阈值Key words
fluorescent magnetic examination/weigh tedtemplate/adaptive neighborhood selection/Ridler adaptive threshold分类
信息技术与安全科学引用本文复制引用
初延亮,肖宇峰,刘桂华,张华..基于荧光磁粉图像的缺陷检测技术[J].微型机与应用,2014,(20):32-34,3.
