基于小波分析的X射线焊接图像中缺陷提取方法  

Approach for Extracting the Defect in the X-Ray Image of Welding Line Based on Tiny Wave Analysis

作　　者：孙正[1] 张晓光[1] 张燕超[1] 

机构地区：[1]中国矿业大学机电工程学院,江苏徐州221008

出　　处：《矿山机械》2007年第4期108-110,共3页Mining & Processing Equipment

摘　　要：针对射线检测焊接图像内缺陷的提取,本文给出了一种利用小波分析对焊接图像中列灰度曲线奇异点检测进行缺陷提取的方法。把每一列灰度值曲线的微分曲线作为被分析函数f(x),根据Lipschitz指数与小波变换模极大值的关系,利用小波模极大值处检测出曲线f(x)上的所有奇异点,并利用Lipschitz指数来估算边缘点奇异程度,由此实现焊缝和缺陷边缘的精确检测。仿真结果表明,利用小波分析方法可取得较好的焊缝、缺陷边缘提取效果。According to the defect extraction from the X-ray image of welding line, the paper offers a kind of approach for detecting the special point on the list grey curve and extracting the defect based on tiny wave analysis. The differential curve of each grey curve serving as the analyzed function f (x), according to the relationship between Lipschitz exponential and the maximum of tiny wave alternation module, which is applied to detect all the special point on the curve f (x), Lipschitz exponential is applied to estimate the special degree of fringe point, thus, the precious detection to the welding line and the defect fringe is realized. The simulation result shows that the tiny wave analysis approach could be applied to obtain the better effect of extracting the defect on welding line and defect fringe.

关 键 词：小波分析 奇异点 边缘检测 缺陷提取 

分 类 号：TG441.7[金属学及工艺—焊接]