Parametric Study of Anomaly Detection Models for Defect Detection in Infrared Thermography

IF 0.9 4区 材料科学 Q4 MATERIALS SCIENCE, CHARACTERIZATION & TESTING Russian Journal of Nondestructive Testing Pub Date : 2024-02-20 DOI:10.1134/S1061830923600600
G. T. Vesala, V. S. Ghali, Y. Naga Prasanthi, B. Suresh
{"title":"Parametric Study of Anomaly Detection Models for Defect Detection in Infrared Thermography","authors":"G. T. Vesala,&nbsp;V. S. Ghali,&nbsp;Y. Naga Prasanthi,&nbsp;B. Suresh","doi":"10.1134/S1061830923600600","DOIUrl":null,"url":null,"abstract":"<p>In the current NDT 4.0 revolution, machine learning and artificial intelligence have emerged as the major enablers for non-destructive testing and evaluation (NDT&amp;E) of industrial components. However, recent developments in active thermal NDT (TNDT) support its use as a practical method for checking a range of industrial components. Additionally, recent post-processing research in TNDT has developed several machine learning models to replace human interaction and offer automatic defect detection. However, the smaller area of the flaws and their related few thermal profiles than the wide sound area, leading to imbalanced datasets, make it difficult to train a supervised deep neural. Recently added to TNDT are anomaly detection models and one-class classifiers, both of which are commonly applied machine learning models to real-world issues. The accuracy and other important metrics in autonomous defect detection are influenced by the hyper-parameters of these models, such as contamination factor, volume of training data, and initialization parameter of the relevant model. The current paper investigates how initialization parameters affect these models' TNDT capabilities for automated flaw detection. Using quadratic frequency modulated thermal wave imaging (QFMTWI), a carbon fiber-reinforced polymer specimen with variously sized artificially produced back-holes at different depths is examined. A good hyper-parameter for automatic flaw identification is chosen after qualitatively comparing testing accuracy, precision, recall, F-score, and probability.</p>","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":null,"pages":null},"PeriodicalIF":0.9000,"publicationDate":"2024-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Russian Journal of Nondestructive Testing","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1134/S1061830923600600","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"MATERIALS SCIENCE, CHARACTERIZATION & TESTING","Score":null,"Total":0}
引用次数: 0

Abstract

In the current NDT 4.0 revolution, machine learning and artificial intelligence have emerged as the major enablers for non-destructive testing and evaluation (NDT&E) of industrial components. However, recent developments in active thermal NDT (TNDT) support its use as a practical method for checking a range of industrial components. Additionally, recent post-processing research in TNDT has developed several machine learning models to replace human interaction and offer automatic defect detection. However, the smaller area of the flaws and their related few thermal profiles than the wide sound area, leading to imbalanced datasets, make it difficult to train a supervised deep neural. Recently added to TNDT are anomaly detection models and one-class classifiers, both of which are commonly applied machine learning models to real-world issues. The accuracy and other important metrics in autonomous defect detection are influenced by the hyper-parameters of these models, such as contamination factor, volume of training data, and initialization parameter of the relevant model. The current paper investigates how initialization parameters affect these models' TNDT capabilities for automated flaw detection. Using quadratic frequency modulated thermal wave imaging (QFMTWI), a carbon fiber-reinforced polymer specimen with variously sized artificially produced back-holes at different depths is examined. A good hyper-parameter for automatic flaw identification is chosen after qualitatively comparing testing accuracy, precision, recall, F-score, and probability.

Abstract Image

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于红外热成像缺陷检测的异常检测模型参数研究
摘要 在当前的无损检测 4.0 革命中,机器学习和人工智能已成为工业部件无损检测和评估(NDT&E)的主要推动因素。然而,主动热无损检测(TNDT)的最新发展也支持将其作为检测一系列工业部件的实用方法。此外,最近在 TNDT 方面的后处理研究已经开发出多个机器学习模型,以取代人工互动,提供自动缺陷检测。然而,与宽广的声音区域相比,缺陷的面积较小,与之相关的热剖面较少,导致数据集不平衡,因此很难训练有监督的深度神经。最近,TNDT 增加了异常检测模型和单类分类器,这两种模型都是实际问题中常用的机器学习模型。自主缺陷检测的准确性和其他重要指标受这些模型的超参数影响,如污染因子、训练数据量和相关模型的初始化参数。本文研究了初始化参数如何影响这些模型在自动缺陷检测中的 TNDT 能力。通过使用二次频率调制热波成像(QFMTWI),研究了在不同深度上具有不同大小人工制造的背孔的碳纤维增强聚合物试样。在定性比较了测试准确度、精确度、召回率、F 分数和概率之后,选择了一个用于自动识别缺陷的良好超参数。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Russian Journal of Nondestructive Testing
Russian Journal of Nondestructive Testing 工程技术-材料科学:表征与测试
CiteScore
1.60
自引率
44.40%
发文量
59
审稿时长
6-12 weeks
期刊介绍: Russian Journal of Nondestructive Testing, a translation of Defectoskopiya, is a publication of the Russian Academy of Sciences. This publication offers current Russian research on the theory and technology of nondestructive testing of materials and components. It describes laboratory and industrial investigations of devices and instrumentation and provides reviews of new equipment developed for series manufacture. Articles cover all physical methods of nondestructive testing, including magnetic and electrical; ultrasonic; X-ray and Y-ray; capillary; liquid (color luminescence), and radio (for materials of low conductivity).
期刊最新文献
Erratum to: Analysis of Weak Signal Detection Based on Tri-Stable System under Poisson White Noise Nondestructive Detection of Wire Rope Damage Using Leakage Magnetic Technique based on Dual-Layer Sensors Erratum to: Solid Particle Erosion Behaviour of Laser Sintered Heat Treated Ti–6Al–4V Alloy Enhanced Electromagnetic Near Field Probe for Diagnosis and Materials Characterization Some Cases of Explicit Expression of the Intensity of the Resulting Field of Magnets Placed in the Field of External Sources
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1