作者：邱根, 王锂, 陈凯
作者单位：电子科技大学自动化工程学院，四川 成都 611731
关键词：高压容器;无损检测;SURF算法;MSAC算法;图像融合
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。
Research on high pressure vessel detection technology based on infrared image fusion algorithms
QIU Gen, WANG Li, CHEN Kai
School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu 611731,China
Abstract: 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.
Keywords: high pressure vessel;nondestructive testing;SURF algorithm;MSAC algorithm;image fusion
2021, 47(5):97-103 收稿日期: 2021-01-31;收到修改稿日期: 2021-02-23
基金项目: 四川省重大科学仪器设备专项（2019ZDZX0045）; 中央高校基本业务费（ZYGX2019J063）
作者简介: 邱根(1982-),四川乐山市人,工程师,博士研究生,主要从事宽带时域测试、功率测量与能效分析、电力测控设备及仪器开发等
参考文献
[1] 范志超, 陈学东, 崔军, 等. 我国重型压力容器轻量化设计制造技术研究进展[J]. 压力容器, 2013, 30(2): 59-65
[2] PROCZKA J, MURALIDHARAN K, VILLELA D, et al. Guidelines for the pressure and efficient sizing of pressure vessels for compressed air energy storage[J]. Energy Conversion & Management, 2013, 65: 597-605
[3] 薛飞龙. 钢制低温压力容器检验及安全评估[J]. 化工设计通讯, 2019, 45(7): 219
[4] 唐永进. 压力管道应力分析的内容及特点[J]. 石油化工设计, 2008(2): 20-24
[5] 占杰龙. 压力容器焊接技术及质量缺陷分析[J]. 化工设计通讯, 2019, 45(7): 93-94
[6] TOMÁŠ I, VÉRTESY G, GILLEMOT F, et al. Nondestructive magnetic adaptive testing of nuclear reactor pressure vessel steel degradation[J]. Journal of Nuclear Materials, 2013, 432(1-3): 371-377
[7] 沈功田, 张万岭. 压力容器无损检测技术综述[J]. 无损检测, 2004(1): 37-40
[8] HUANG X, YIN C, DADRAS S, et al. Adaptive rapid defect identification in ECPT based on K-means and automatic segmentation algorithm[J]. Journal of Ambient Intelligence and Humanized Computing, 2017: 1-18
[9] ZHANG B, CHENG Y H, YIN C, et al. Design of an automatic defect identification method based ECPT for pneumatic pressure equipment[J]. Complexity, 2018, ArticleID 5423924, https://doi.org/10.1155/2018/5423924.
[10] BAI L B, CHENG Y H, CHEN Y F, et al. ICA fusion approach based on fuzzy using in eddy current pulsed thermography[J]. International Journal of Applied Electromagnetics and Mechanics, 2016, 52(1-2): 443-451
[11] 郑兰, 安博文, 曹芳. 一种基于特征点匹配的红外图像拼接算法[J]. 计算机应用与软件, 2015, 32(9): 192-196
[12] BAY H, TUYTELAARS T, GOOL L V. SURF: Speeded up robust features[C]//Proceedings of the 9th European Conference on Computer Vision, 2006.
[13] WANG R, SHI Y, CAO W. GA-SURF: A new speeded-up robust feature extraction algorithm for multispectral images based on geometric algebra[J]. Pattern Recognition Letters, 2019, 127: 11-17
[14] TORR P, ZISSERMAN A. MLESAC: A new robust estimator with application to estimating image geometry[J]. Computer Vision and Image Understanding, 2000, 78(1): 138-156
[15] 曲天伟, 安波, 陈桂兰. 改进的RANSAC算法在图像配准中的应用[J]. 计算机应用, 2010, 30(7): 1849-1851
[16] 徐鑫, 孙韶媛, 沙钰杰, 等. 一种基于改进RANSAC的红外图像拼接方法[J]. 激光与光电子学进展, 2014, 51(11): 135-140
[17] 王晓宇. 红外图像分析关键技术研究[D]. 武汉: 华中科技大学, 2008.
[18] HEMMAT H J, POURTAHERIAN A, BONDAREV E, et al. Fast planar segmentation of depth images[C]//Processing of SPIE- the International Society for Optics and Photonics, 2015.
