Suzhou Electric Appliance Research Institute
期刊號： CN32-1800/TM| ISSN1007-3175

Article retrieval

文章檢索

首頁 >> 文章檢索 >> 最新索引

用于電力設備異常診斷的圖像配準及融合方法

來源：電工電氣發(fā)布時間：2024-12-02 08:02 瀏覽次數(shù)：30

用于電力設備異常診斷的圖像配準及融合方法

周喜紅1，席亞賓1，李中寶2
(1 廣東粵電大亞灣綜合能源有限公司，廣東 惠州 516000；
2 中國核工業(yè)二三建設有限公司，北京 101300)
 
    摘 要：近年來圖像融合方法在電力設備熱異常的診斷中所占比重逐漸增加，但是涉及到圖像配準和融合統(tǒng)一考慮的方法很少。提出了一種最大迭代關聯(lián)圖像配準及區(qū)域特性判別的圖像融合方法，用于輔助熱異常的診斷。該方法通過構(gòu)建約束函數(shù)計算源圖像配準迭代次數(shù)，隸屬度函數(shù)定義源圖像的區(qū)域特性，已知區(qū)域特性的子圖像根據(jù)電力設備熱異常所重視的特征優(yōu)先選擇融合策略，以最大程度保留源圖像中的紋理特征和熱輻射特征。在自建的電力設備數(shù)據(jù)集上與其他方法對比顯示，所提方法在保證源圖像配準精度的前提下，還突出了紅外圖像的熱輻射特征和可見光圖像的紋理特征，能夠滿足電力設備熱異常診斷的需要。
    關鍵詞： 圖像融合；圖像配準；電力設備；熱異常診斷；約束函數(shù)；隸屬度函數(shù)；熱輻射
    中圖分類號：TM711 ；TP391     文獻標識碼：B     文章編號：1007-3175(2024)11-0067-10
 
Image Registration and Fusion Method for Anomaly
Diagnosis of Power Equipment
 
ZHOU Xi-hong1, XI Ya-bin1, LI Zhong-bao2
（1 Guangdong Yuedian Daya Bay Integrated Energy Co., Ltd, Huizhou 516000, China;
2 China Nuclear Industry 23 Construction Co., Ltd, Beijing 101300, China）
 
    Abstract: In recent years, the proportion of image fusion methods in the diagnosis of thermal anomalies of power equipment has gradually increased, but the methods involving unified consideration of image registration and fusion are rare. Therefore, this paper proposes an image fusion method based on maximum iterative correlation image registration and regional feature discrimination, which is used to assist thermal anomaly diagnosis. This method calculates the number of source image registration iterations by constructing a constraint function, and the membership function defines the regional characteristics of the source image. The sub-images with known regional characteristics preferentially select the fusion strategy according to the characteristics that the thermal anomaly of the power equipment attaches importance to, so as to retain the texture features and thermal radiation features in the source image to the greatest extent. Compared with other methods on the self-built power equipment dataset, the proposed method not only ensures the registration accuracy of the source image, but also highlights the thermal radiation characteristics of the infrared image and the texture characteristics of the visible image, which can meet the needs of thermal anomaly diagnosis of power equipment.
    Key words: image fusion; image registration; power equipment; thermal anomaly diagnosis; constraint function; membership function;thermal radiation
 
參考文獻
[1] CHEN Xiaolong, WANG Peihong, HAO Yongsheng, et al.Evidential KNN-Based Condition Monitoring and Early Warning Method with Applications in Power Plant[J].Neurocomputing，2018，315 ：18-32.
[2] HUANG Z, XIE W, LIU W, et al.TSCDNet +: A Highly Robust Substation Anomaly Detection Method[J].Optik，2021，246 ：167808.
[3] NAN L D, RUI H, QIANG L, et al.Research on Fuzzy Enhancement Algorithms for Infrared Image Recognition Quality of Power Internet of Things Equipment Based on Membership Function[J].Journal of Visual Communication & Image Representation，2019，62 ：359-367.
[4] ZOU H, HUANG F.A Novel Intelligent Fault Diagnosis Method for Electrical Equipment Using Infrared Thermography[J].Infrared Physics & Technology，2015，73 ：29-35.
[5] 魯曉涵，李洋，邰昱博，等. 基于 GAN 輕量化改進的紅外與可見光圖像融合算法[J] . 電光與控制，2024，31(8) ：58-62.
[6] 馮新文，劉璟明，朱呂甫. 基于 MSR 和 BCI 的變電站巡檢圖像融合方法[J] . 電力信息與通信技術，2022，20(4) ：94-101.
[7] 陰錫君，劉郁，王一珺. 圖像融合技術在變電站設備熱故障監(jiān)測中的應用研究[J] . 科技通報，2019，35(12) ：121-124.
[8] JIANG Qian, LIU Yadong, YAN Yingjie, et al.A Contour Angle Orientation for Power Equipment Infrared and Visible Image Registration[J].IEEE Transactions on Power Delivery，2020，36(4) ：2559-2569.
[9] LU Mingshu, LIU Haiting, YUAN Xipeng.Thermal Fault Diagnosis of Electrical Equipment in Substations Based on Image Fusion[J]. Traitement Du ：Signal Imageparole，2021，38(4) ：1095-1102.
[10] 李健，王濱海，李麗，等. 基于 SIFT 的電力設備紅外與可見光圖像的配準和融合[J] . 光學與光電技術，2012，10(1) ：75-78.
[11] XU Han, MA Jiayi, YUAN Jiteng, et al.RFNet:Unsupervised Network for Mutually Reinforcing Multi-Modal Image Registration and Fusion[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)，2022.
[12] ZHU Qidan, JING Liqiu, BI Rongsheng.Exploration and Improvement of Ostu Threshold Segmentation Algorithm [C]//The 8th World Congress on Intelligent Control and Automation, 2010.
[13] JING Zhongliang.Image Fusion Based on an Expectation Maximization Algorithm[J].Optical Engineering，2005，44(7) ：077001.
[14] PALSSON F, SVEINSSON J R, ULFARSSON M O,et al . Model-Based Fusion of Multi-and Hyperspectral Images Using PCA and Wavelets[J].IEEE Transactions on Geoscience & Remote Sensing，2015，53(5) ：2652-2663.
[15] SHEN R, CHENG I, BASU A.Cross-Scale Coefficient Selection for Volumetric Medical Image Fusion[J].IEEE Transactions on BiomedicalEngineering，2012，60(4) ：1069-1079.
[16] JAGER F, HORNEGGER J.Nonrigid Registration of Joint Histograms for Intensity Standardization in Magnetic Resonance Imaging [J] . IEEE Transactions on Medical Imaging，2008，28(1) ：137-150.
[17] GONCALVES H, CORTE-REAL L, GONCALVES J A.Automatic Image Registration Through Image Segmentation and SIFT[J].IEEE Transactions on Geoscience and Remote Sensing，2011，49(7) ：2589-2600.
[18] BAY H, ESS A, TUYTELAARS T, et al.Speeded-Up Robust Features(SURF)[J].Computer Vision & Image Understanding，2008，110(3) ：346-359.
[19] LI Shutao, KANG Xudong, HU Jianwen.Image Fusion with Guided Filtering[J].IEEE Transactions on Image Processing，2013，22(7) ：2864-2875.
[20] MA J Y, CHEN C, LI C, et al.Infrared and Visible Image Fusion Via Gradient Transfer and Total Variation Minimization[J].Information Fusion，2016，31 ：100-109.
[21] BAVIRISETTI D P, XIAO G, LIU G.Multi-Sensor Image Fusion Based on Fourth Order Partial Differential Equations[C]//2017 20th International Conference on Information Fusion，2017.
[22] YAN Lei, CAO Jie, RIZVI Saad, et al.Improving the Performance of Image Fusion Based on Visual Saliency Weight Map Combined with CNN[J].IEEE Access，2020，8 ：59976-59986.
[23] XU Han, MA Jiayi, JIANG Junjun, et al.U2Fusion:A Unified Unsupervised Image Fusion Network[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2022，44(1) ：502-518.
[24] MA Jiayi, XU Han, JIANG Junjun, et al.DDcGAN:A Dual-Discriminator Conditional Generative Adversarial Network for Multi-Resolution Image Fusion[J].IEEE Transactions on Image Processing，2020，29 ：4980-4995.
[25] MA Jiayi, ZHANG Hao, SHAO Zhenfeng, et al.GANMcC:A Generative Adversarial Network with Multiclassification Constraints for Infrared and Visible Image Fusion[J].IEEE Transactions on Instrumentation and Measurement，2020，70 ：5005014.
[26] MA Jiayi, TANG Linfeng, XU Meilong, et al.STDFusionNet:An Infrared and Visible Image Fusion Network Based on Salient Target Detection[J].IEEE Transactions on Instrumentation and Measurement，2021，70 ：5009513.
[27] SENGUPTA D, GUPTA P, BISWAS A.A Survey on Mutual Information Based Medical Image Registration Algorithms[J].Neurocomputing，2021，486 ：174-188.
[28] ROBERTS J W, AARDT J V, AHMED F.Assessment of Image Fusion Procedures Using Entropy, Image Quality, and Multispectral Classification[J].Journal of Applied Remote Sensing，2008，2(1) ：1-28.
[29] WANG E , YANG B , PANG L . Superpixel-Based Structural Similarity Metric for Image Fusion Quality Evaluation[J].Sensing and Imaging，2021，22(1) ：1-25.
[30] XYDEAS C S , PV V . Objective Image Fusion Performance Measure[J].Military Technical Courier，2000，56(4) ：181-193.
[31] ESKICIOGLU A M, FISHER P S.Image Quality Measures and Their Performance [J] . IEEE Transactions on Communications，1995，43(12) ：2959-2965.