基于红外图像融合算法的高压容器检测技术研究  

Research on high pressure vessel detection technology based on infrared image fusion algorithms

作　　者：邱根[1] 王锂 陈凯[1] QIU Gen;WANG Li;CHEN Kai(School of Automation Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China)

机构地区：[1]电子科技大学自动化工程学院,四川成都611731

出　　处：《中国测试》2021年第5期97-103,共7页China Measurement & Test

基　　金：四川省重大科学仪器设备专项(2019ZDZX0045);中央高校基本业务费(ZYGX2019J063)。

摘　　要：针对红外热成像技术对高压容器表面缺陷进行无损检测过程中,由于其体积庞大而无法实现对容器整体单次完全检测的情况以及单幅图像可能无法展示缺陷整体形状和大小的问题,大量学者和研究人员对高压容器的无损检测方法进行了研究。该文采用加速稳健特征(SURF)算法对多次检测得到红外图像进行特征提取,通过双向匹配法和M估计样本一致(MSAC)算法进行特征点对匹配,并提出一种基于亮度和距离的加权融合算法对参考图像和待配准图像进行融合。最后,通过仿真实验,对高压容器复合金属板的362帧红外采集样本图像进行特征提取和融合处理,实验结果可证明算法对缺陷信息检测和拼接融合显示的有效性。In the process of non-destructive testing for surface defects of high-pressure containers by infrared thermal imaging technology,due to its large size,it is impossible to achieve a single complete detection of the whole container,and a single image may not be able to show the overall shape and size of the defects.In recent years,a large number of scholars and researchers have studied the detection methods of high-pressure containers.In this paper,the speeded up robust features(SURF)algorithm is used to extract the features of the infrared image,and use the two way matching method and the M-estimator sample consensus(MSAC)algorithm to match the feature points.A weighted fusion algorithm based on brightness and distance is proposed to fuse the reference image and the image to be registered.Finally,through the simulation experiment,the feature of 362 infrared sample images of high-pressure containers composite metal plate are extracted and the images are fused.Experimental results demonstrate the effectiveness of the algorithm in defect information detection and fusion display.

关 键 词：高压容器 无损检测 SURF算法 MSAC算法 图像融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]