X射线无损检测图像质量提升算法  

Quality improvement algorithm for x-ray non-destructive testing image

作　　者：梅健强[1] 陈菀妍 贾凡 MEI Jianqiang;CHEN Wanyan;JIA Fan(School of Electronic Engineering,Tianjin University of Technology and Education,Tianjin 300222,China;Raysov Instrument Co.Ltd.,Dandong 118017,China)

机构地区：[1]天津职业技术师范大学电子工程学院,天津300222 [2]丹东锐新射线仪器有限公司,丹东118017

出　　处：《天津职业技术师范大学学报》2023年第1期44-49,共6页Journal of Tianjin University of Technology and Education

基　　金：天津市科技计划项目(22YDTPJC00720);教育部协同育人项目(202102186003).

摘　　要：针对X射线工业无损检测(NDT)中,物体材质形状各异造成图像对比度低影响缺陷检测精度的问题,基于改进的视网膜Retinex理论提取图像关键信息,及对比度受限自适应直方图均衡(CLAHE)和伽玛映射增强图像对比度和亮度,使用非锐化掩模(USM)循环强化边缘。定性和定量分析表明,研究所提方法可以在增强图像对比度的同时锐化边缘,达到提升X射线无损检测图像质量的目的。Low contrast due to the different materials and shapes of objects affects the accuracy of defect inspection in industrial X-ray non-destructive testing(NDT). The crucial information of the image is extracted based on the improved Retinex theory, and the contrast and brightness of the image are enhanced based on contrast limited adaptive histogram equalization(CLAHE)and gamma mapping, the edge information is promoted through a cycle of unsharp mask(USM). Qualitative and quantitative analysis shows that enhanced contrast with sharpening edge effects could be obtained through the proposed method, which improves the image quality of X-ray NDT.

关 键 词：图像质量提升 图像增强 无损检测 X射线成像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]